Clinical Breast Imaging: A Patient Focused Teaching File, 1st Edition

Chapter 4 - Management

 

Terms

·     Atypical ductal hyperplasia (ADH)

·     Atypical lobular hyperplasia (ALH)

·     Columnar alteration with prominent apical snouts and secretions (CAPSS)

·     Complex sclerosing lesion

·     Cribriform ductal carcinoma in situ

·     Ductal carcinoma in situ (DCIS)

·     Ductography

·     E-cadherin

·     False negative (FN)

·     False positive (FP)

·     Fibroadenoma

·     Hyperplasia

·     Incident cancer detection rate

·     Intracystic carcinoma

·     Ipsilateral breast tumor recurrence (IBTR)

·     Lobular carcinoma in situ (LCIS)

·     Lobular neoplasia

·     Local recurrence

·     Medical audit

·     Micropapillary ductal carcinoma in situ

·     Minimal breast cancer

·     Mucocele-like lesion

·     Multiple peripheral papillomas

·     Papilloma

·     Papillomatosis

·     Phyllodes tumor

·     Pneumocystogram

·     Positive predictive value (PPV)

·     Prevalent cancer detection rate

·     Pseudoangiomatous stromal hyperplasia (PASH)

·     Radial scar

·     Regional recurrence

·     Sclerosing adenosis

·     Secretory carcinoma

·     Sensitivity

·     Specificity

·     True negative (TN)

·     True positive (TP)

Management

In managing the various situations that arise in breast imaging, it is good to always be thinking not just relative to what the first step should be for your patient's care, but what the second and third steps might be as well. Your patient? Yes, your patient. Know and consider the cascade of events you precipitate for patients based on what you say to them, how you word your report, and the recommendations you make. Are you sure enough about what you are saying to justify whatever ensues for the patient? Is your decision motivated by a defensive posture and the recognition, at some level, that the workup is incomplete (substandard) and that you are operating with inadequate or incomplete information, or one that is justifiably based on common sense, a complete workup, and what is good for the patient?

Correlation

The need for correlation in every process undertaken is fundamentally important. If an ultrasound is done to evaluate a mammographic or magnetic resonance imaging (MRI) finding, does what you see on ultrasound correlate with the mammographic or MRI finding? If the patient presents with a focal finding, does what you see mammographically correlate with the described clinical finding? When recommending an imaging-guided or excisional biopsy, consider what you will accept as a diagnosis and what you will recommend if the results are different from those expected. Are the imaging and histologic findings concordant? If the imaging and histologic findings are benign and congruent, patients can be returned to annual screening mammography. For patients diagnosed with a malignancy, MRI and surgical consultation are scheduled. If the findings are not concordant, repeat biopsy or excisional biopsy is recommended. For patients with a diagnosis of atypical ductal hyperplasia (ADH), possible phyllodes tumor, multiple peripheral papillomas, fibromatosis, or granular cell tumor, wide surgical excision is indicated following imaging-guided biopsies.

The management of several types of lesions diagnosed on core biopsies remains controversial. Included in this group are solitary papillomas, lobular neoplasia (atypical lobular neoplasia, lobular carcinoma in situ), complex sclerosing lesions, and mucocele-like lesions. The current consensus is that excisional biopsy is appropriate when these lesions are diagnosed on core biopsies because of the reported incidence of associated malignancy and the frequency with which some of these lesions are upgraded to cancer when more tissue is examined histologically.

Breast Imaging Consultations

As mentioned previously, regardless of the type of study done, the title on the written description of what I do is worded as a “Breast Imaging Consultation” not a “Radiology Report.” The comprehensive evaluations that can be undertaken by breast imagers have put us in a consultative role. A general outline used for breast imaging consultative reports includes:

·     Type of study (e.g., screening or diagnostic mammogram, ultrasound)

·     Succinct description and location of findings

·     Impression, with your specific recommendations

·     Assessment category (required under the Mammography Quality Standards Act for all mammographic studies) with wording as provided by the American College of Radiology, Breast Imaging and Reporting Data System (BI-RADS®) for mammography:

·     Category 1: negative

·     Category 2: benign finding

·     Category 3: probably benign finding; short-interval follow-up is recommended

·     Category 4: suspicious abnormality—biopsy should be considered

·     Category 5: highly suggestive of malignancy—appropriate action should be taken

·     Category 6: known biopsy-proven malignancy—appropriate action should be taken

·     Category 0: need additional imaging evaluation and/or prior mammograms for comparison

·     Category 4 lesions can be further subdivided at the discretion of the facility for their internal use into:

·     Category 4A: low suspicion for malignancy

·     Category 4B: intermediate suspicion for malignancy

·     Category 4C: moderate concern

I make every effort to generate descriptive but succinct reports that are clinically relevant and that provide specific direction and recommendations. I use no disclaimers in my reports, and I do not abdicate clinical correlation of anything to others. In considering the wording of reports, it is my contention that if, based on complete clinical and imaging evaluations, you have a high degree of certainty relative to the diagnosis, clear consultative reports with specific recommendations can be dictated easily and succinctly (e.g., “a spiculated mass measuring 7 mm is imaged at the 3 o'clock position, 5 cm from the right nipple. Biopsy is indicated. This is undertaken and reported separately.”). Make up your mind about what you are going to say before you start dictating. Hedges and disclaimers are used when we are uncomfortable and uncertain about a finding and its significance. In this situation, I would suggest that we need to do whatever it takes to increase our level of certainty so that we can be more definitive.

For potentially abnormal screening mammograms, I state what the potential abnormality is (e.g., mass, calcifications, distortion) and in which breast it is located; I also comment specifically on the other breast. The description and characterization of the lesion is deferred until a thorough evaluation is completed (e.g., for a mass, spot compression views, physical examination, and an ultrasound). The recommendation for a woman with a potentially abnormal screening mammogram is “additional evaluation is indicated and we will contact the patient directly to schedule the additional evaluation” (see the introduction to Chapter 2 for additional discussion). Consequently, the only assessment categories used for screening mammograms are 0, 1, and 2.

In the diagnostic setting, I issue one consultative report that includes the findings for the diagnostic mammogram, physical examination, and ultrasound. Based on the clinical and imaging features of a lesion, the findings are described and an impression with recommendations is generated. The impression is not used to repeat a description of the findings but rather is used to deliver the final, clinically relevant concept with what I think is indicated for the patient.

Read your reports critically. Strive for precision (e.g., give measurements for a lesion and avoid characterizations such as “small” or “large”) and eliminate unnecessary words (e.g., “clearly,” “appears to be,” “very,” etc.) that provide no relevant information and may serve to obscure your message. It is important to familiarize yourself with the mammography, ultrasound, and magnetic resonance imaging lexicons provided by the American College of Radiology, Breast Imaging and Reporting Data System (BI-RADS®), as these provide guidelines on what should be described for particular findings and suggests terminology to be used for relevant findings.

Medical Audit

In breast imaging, accountability needs to be present every step of the way. Although quality control and data tracking are often relegated to the technologist, radiologists should be actively involved in these processes for their practice. It is only through monitoring results that problems can be addressed and much learning can take place. We need to know how well we are doing, and we want to identify potential problems that can be addressed so that patient care is improved. By tracking data and learning from the results, we can improve patient care. The numbers generated from the audit should be viewed not as one point in time but rather how they change as you gain more experience.

Data that should be collected include:

·     Date of audit

·     Number of screening studies (first-time study vs. repeat screen)

·     Number of diagnostic studies

·     Call-back (recall) recommendations (e.g., BI-RADS® category 0: need additional imaging evaluation)

·     Biopsy recommendations (BI-RADS® category 4 and 5: suspicious abnormality and highly suggestive of malignancy)

·     Biopsy results (e.g., benign vs. malignant; FNA vs. core biopsy vs. excisional biopsy)

·     Tumor staging: histologic subtype, grade, size, and nodal status

Data that you can calculate include:

·     True positive (TP): cancer diagnosed within 1 year of a biopsy recommendation for an abnormal mammogram

·     True negative (TN): no known cancer within 1 year of a normal mammogram

·     False negative (FN): cancer diagnosed within 1 year of a normal mammogram; these should be reviewed and analyzed

·     False positive (FP):

·     FP1 = no known cancer diagnosed within 1 year of an abnormal screening mammogram for which additional imaging or biopsy is recommended

·     FP2 = no known cancer diagnosed within 1 year of a recommendation for biopsy or surgical consultation based on an abnormal mammogram

·     FP3 = benign disease diagnosed on biopsy within 1 year after recommendation for biopsy or surgical consultation based on an abnormal mammogram

·     Positive predictive value:

·     PPV1 = percentage of cancers diagnosed following an abnormal screening mammogram

·     PPV2 = percentage of cancers diagnosed when a biopsy or surgical consultation is recommended following a screening mammogram

·     PPV3 = percentage of cancers diagnosed on the actual number of biopsies done as a result of a screening mammogram

·     Cancer detection rate for asymptomatic women (i.e., true screening population)

·     Prevalent (rate of cancer detection among women presenting for their first screening mammogram)

·     Incident (rate of cancer detection among women with prior screening mammograms)

·     By age groups

·     Percentage of minimal breast cancers diagnosed

·     Percentage of node-negative breast cancers diagnosed

·     Call-back (recall) rate

·     Sensitivity =TP/(TP + FN), or the probability of detecting a cancer when a cancer is present

·     Specificity =TN/(FP + TN), or the probability of a normal mammogram when no cancer is present

Based on reports of audit data, obtainable goals include:

·     PPV1 (abnormal screens): 5% to 10%

·     PPV2 (biopsy recommendations): 25% to 40%

·     Stage 0 or 1 tumors diagnosed: > 50%

·     Minimal cancers (invasive cancer ≥ 1 cm or DCIS): > 30%

·     Node-positive tumors: < 25%

·     Prevalent cancers/1,000: 6 to 10

·     Incident cancers/1,000: 2 to 4

·     Call-back (recall) rate: ≥10%

·     Sensitivity: > 85%

·     Specificity: > 90%

Published sensitivity rates for mammography range between 85% and 90%. Recognize, however, that this is probably one of the harder statistics to obtain because of the difficulty of establishing an accurate false negative rate. Access to a statewide tumor registry can be helpful; however, if the patient moves (or seeks medical care) out of state, knowledge of a cancer diagnosis may not be readily accessible to the screening facility.

The effect of breast imaging and the role of radiologists in the management of women with breast cancer goes unstated and, in many ways, is often misrepresented. There is continued skepticism and criticisms relative to our contributions to patient care and the significance of what has already been accomplished: the routine identification of lymph node-negative stage 0 and stage I invasive cancers and ductal carcinoma in situ. Recently reported decreases in breast cancer mortality rates are attributed by many to more effective treatment, ignoring or relegating to a secondary role our ability to detect DCIS, stage 0, and stage I lesions in many patients. Is early detection possibly the more important factor, and does not our ability to identify small lesions increase available treatment options and render them more effective for patients?

 

Patient 1

Figure 4.1. Screening study, 60-year-old woman. Craniocaudal (A) and mediolateral oblique (B) views, right breast. Craniocaudal (C) and mediolateral oblique (D) spot compression views, right breast.

What do you think?

What would you do next?

A mass with a radiolucent center, associated distortion, and long spicules is present anteriorly in the upper outer quadrant of the right breast. Given these features, fat necrosis related to a prior biopsy is one of the main considerations. As a starting point, prior films would be helpful in determining if this finding could be seen previously, and what, if any, evolution has occurred. A review of the patient's history form, looking specifically to see if she has had a biopsy at this site, will also be helpful. Unfortunately, no prior films are available for this patient, and nothing is indicated on the history form relative to a prior biopsy in the right breast. Additional evaluation is indicated.

BI-RADS® category 0: need additional imaging evaluation.

The spot compression views confirm the presence of a mass with a low-density central area, distortion, and long spiculation. The differential for this finding includes fat necrosis related to prior surgery or trauma, complex sclerosing lesion, sclerosing adenosis, papilloma, focal fibrosis, an inflammatory process, invasive ductal carcinoma not otherwise specified, tubular carcinoma, and invasive lobular carcinoma. In discussing the findings with the technologist, she relates not seeing a scar at the site of the spot views. Correlative physical examination and an ultrasound are indicated next.

After introducing myself to the patient and briefly describing what we have seen so far and what I would like to do next, I specifically ask her about prior breast surgery or trauma. She has no recollection of having had surgery or trauma to the right breast. On close inspection of the periareolar region, however, a scar is apparent at the edge of the areola corresponding to the site of the mammographic finding. I palpate no corresponding mass, elicit no tenderness at this site, and the ultrasound is normal throughout the subareolar area extending into the upper outer quadrant of the right breast.

As I approached the patient, I remained open to all diagnostic possibilities; however, I had some skepticism relative to the information provided. In my mind, the mammographic findings (low central density, distortion, and long spiculation), in the absence of a palpable finding, were highly suggestive of fat necrosis or a complex sclerosing lesion. So, rather than totally discard my initial impression, I examine the patient carefully, focusing my attention on the site of the mammographic findings. Although the patient has no recollection of prior surgery (patients do not always recall prior surgical procedures, trauma, or inflammatory processes; some may not even recall a breast cancer diagnosis), and the technologist reported no scars, by spending 2 minutes closely examining the patient, the diagnosis of postoperative change is established and no additional intervention or follow-up is recommended.

Given our total fascination with technology and images, we often dismiss or underestimate simple and inexpensive tools such as physical examination, and yet this often provides the correct answer expeditiously. I cannot emphasize enough how many times direct communication with the patient and a thorough physical examination provide an efficient means of arriving at the correct answer for a given patient. As clinical breast imagers, we are in a unique position to integrate clinical, physical, imaging, and histologic findings to provide accurate, optimal patient care. Do not sell yourself or your patients short by passing up an opportunity to examine and talk to the patient directly. Do not relegate physical examination and the performance of ultrasound studies to others.

 

Patient 2

Figure 4.2. Diagnostic evaluation, 33-year-old patient presenting with a “lump” in the right breast. Craniocaudal (A) and mediolateral oblique (B) views, metallic BB used to mark location of “lump” described by the patient. Spot tangential (C) view of the palpable finding, right breast. Ultrasound images, radial (RAD) (D) and antiradial (ARAD) (E) projections, corresponding to the area of concern to the patient in the right breast. Ultrasound image (F), right axilla.

 

What do you think?

What would you do next?

There is dense fibroglandular tissue, and no focal abnormality is apparent on the spot tangential view. Correlative physical examination and an ultrasound are indicated for further evaluation.

On physical examination, a hard, fixed mass is palpated at the 9 o'clock position, 5 cm from the right nipple. On ultrasound, (Fig. 4.2D, E), an irregular mass with a heterogeneous echotexture, indistinct margins, and areas of shadowing and enhancement is imaged corresponding to the palpable finding. This mass measures at least 5 cm. The clinical and sonographic findings are consistent with a malignant process, most likely an invasive ductal carcinoma. A biopsy is indicated.

What else should you do to evaluate this patient further?

For patients in whom we suspect a malignancy, we examine the ipsilateral axilla for potentially abnormal lymph nodes. If any potential abnormality is identified in the axilla, a fine-needle aspiration or a needle biopsy is also done. Patients with metastatic disease to the axilla will undergo a full axillary dissection at the time of the lumpectomy, bypassing the need for a sentinel lymph node biopsy.

In this patient, an oval hypoechoic mass is imaged in the right axilla (Fig. 4.2F). Given the thickening of the cortex and the lack of a hyperechogenic hilar region, this is a potentially abnormal lymph node. Biopsies of the right breast and axillary masses confirm the suspected diagnoses. At the time of her definitive surgery, a grade III invasive ductal carcinoma with extensive lymphovascular space involvement measuring at least 5.7 cm in size is reported histologically. Tumor is also reported in the dermal lymphatics surrounding the right nipple [pT3, pN2, pMX; Stage IIIA].

In patients with palpable findings and normal-appearing dense glandular tissue mammographically, correlative physical examination and an ultrasound are indicated for further evaluation. In this group of patients, ultrasound is an excellent adjunctive tool in evaluating the clinical findings. If our focus is optimal, efficient, and expeditious patient care, then clinical, mammographic, and sonographic evaluations and, when needed, imaging-guided biopsies of the breast and axilla, are done in one visit. With a 24-hour turnaround time on core biopsy results, histology findings are available the following day and the patient is scheduled to see the surgeon for definitive treatment. As clinical breast imagers, we are in a unique position to affect the care our patients receive. Evaluations, which in many communities take weeks, with significant associated anxiety for the patient and her family, can be accomplished accurately in 24 hours. By providing this type of service, we also effectively eliminate the fragmentation of care (and with that the potential for miscommunication) among providers that can result when evaluations are carried out over several weeks (e.g., one radiologist doing the initial mammogram, another doing the ultrasound, and possibly a third doing the biopsies).

 

 

Patient 3

Figure 4.3. Diagnostic evaluation, 38-year-old patient presenting with a palpable mass in the right breast. Craniocaudal (A) and mediolateral oblique (B)views of the right breast. Metallic BB used to mark “lump” described by the patient. Spot compression views of mass in craniocaudal (C) and mediolateral oblique (D) views. Ultrasound images in radial (RAD) (E) and antiradial (ARAD) (F) projections of palpable finding at the 6 o'clock position, approximately 12 cm from the right nipple. Craniocaudal (G) and mediolateral oblique (H) views, right breast, photographically coned. Double spot compression magnification (I) view, right breast.

 

What do you think?

A round, solid, 2-cm mass with spiculated, indistinct, and angular margins and associated shadowing is imaged in the right breast, corresponding to the area of concern to the patient. Although no calcifications are identified associated with the mass on the craniocaudal spot compression view, or on the ultrasound, a cluster of pleomorphic calcifications is evident on the mediolateral oblique spot compression view. Why are these not seen on the spot craniocaudal view? Did you see them on the routine craniocaudal view, or was your eye drawn to the clinical finding? Remember, do not let yourself be distracted by obvious benign or malignant mammographic/clinical findings. Even when you are presented with an obvious finding, make sure to evaluate the remainder of the mammogram and the contralateral side thoroughly.

What do you think now?

What is your recommendation?

The calcifications project on the mass in the mediolateral oblique view, but they are medial in location on the craniocaudal view and at a distance from the mass. On the magnification views, the calcifications are pleomorphic and there are associated linear forms consistent with ductal carcinoma in situ with central necrosis, likely high-nuclear-grade. Biopsies of the mass and calcifications are indicated because if these confirm the suspected diagnoses, this represents multicentric disease and a mastectomy is probably indicated for this patient.

Histologically, a complex ductal carcinoma in situ is diagnosed, corresponding to the cluster of calcifications seen mammographically, and a 2.5-cm, grade III, invasive ductal carcinoma with associated lymphovascular space involvement is described for the mass. Micrometastatic disease is reported in the sentinel lymph node [pT2, pN1mi(sn) (i), pMX, Stage IIB].

Given the incidence of multifocal, multicentric, and bilateral lesions, thorough evaluation of patients who are likely to have a malignancy is critical preoperatively. The mammogram needs to be reviewed carefully. Complete ultrasound evaluations of the breasts and ipsilateral axilla, as well as magnetic resonance imaging, are also helpful in evaluating patients diagnosed with breast cancer. With appropriate imaging protocols, magnetic resonance imaging also makes the evaluation of internal mammary, axillary, supraclavicular, and neck lymph nodes possible for our patients.

Patient 4

Figure 4.4. Screening study, 57-year-old woman. Craniocaudal (A) and mediolateral oblique (B) views. Craniocaudal (C) and mediolateral oblique (D), spot compression views. Ultrasound image, antiradial (ARAD) projection (E), at the eight o'clock position, 4 cm from the left nipple. Diagram (F)illustrating orthogonal images of needle positioning. Obtaining orthogonal images to document final needle positioning is particularly critical when sampling small lesions. The needle may appear to be through the lesion (I); however, on the orthogonal image, the needle is along the edge of the mass (II). Ideally, the needle is surrounded by the lesion on the orthogonal (III) image. Ultrasound images in a different patient demonstrating the needle through the lesion (G) longitudinally. When the transducer is rotated for the orthogonal image (H), the needle is seen at the edge of the mass (arrow) and not through it. Ideally, when the transducer is rotated for the orthogonal image (I), the needle is seen within the mass (arrow). Diagrams that can be used in conjunction with the craniocaudal and mediolateral oblique views to estimate the location of a lesion prior to the ultrasound study. In correlating mammographic and ultrasound findings, it is helpful to start the ultrasound with an estimate of the clock location of a lesion and its expected distance from the nipple. On the mediolateral oblique view/diagram (J), this lesion is “x cm” below the posterior nipple line. On the craniocaudal view/diagram (K), this lesion is medial and posterior in location. The lesion is approximately 8 cm posterior to the nipple. On the frontal diagram (L), which simulates the position of the patient for the ultrasound study, the expected clock position of the lesion is estimated by drawing a line “x cm” below the posterior nipple line. In this patient the lesion should be at 8 o'clock position, 8 cm from the nipple. Ultrasound images, radial (RAD) (M) and antiradial (ARAD) (N)projections obtained at the 8 o'clock position, 8 cm from the left nipple, confirm the location of a mass that corresponds with the mammographic finding. Its sonographic features more closely resemble those of the mass seen mammographically. Single mammographic view (O) obtained following an ultrasound-guided wire localization of the lesion at the 8 o'clock position, 8 cm from the left nipple. The localization wire is through the mass.

 

What do you think?

Is this a normal study, or is there a potential abnormality?

A lucent-centered, benign-type calcification is present in the right breast medially. In reviewing these images systematically, a potential abnormality is noted medially in the left craniocaudal (CC) view. Based on the expected location of this abnormality on the CC view, a low-density nodule is suspected inferiorly on the left mediolateral oblique view. If they are available, prior films would be helpful in assessing whether this finding is new, stable, or decreasing in size. If prior films are not available or this represents a new or enlarging mass, additional evaluation is indicated.

BI-RADS® category 0: need additional imaging evaluation.

What do you think now, and with what level of certainty can you make a recommendation?

Did you notice the fingerprint superimposed on the mass in the craniocaudal spot compression view? This is a plus-density artifact that reflects improper film handling after the film was exposed but before processing. An oval mass with indistinct and spiculated margins is confirmed on the spot compression views. Based on the mammographic findings, this mass requires biopsy. In planning the biopsy, an ultrasound is done because, if the lesion is identified on ultrasound, ultrasound guidance can be used to do the biopsy. If the mass is not identified sonographically, a stereotactically guided biopsy can be done.

BI-RADS® category 4: suspicious abnormality, biopsy should be considered.

Based on the mammographic and sonographic findings, what will you accept as a diagnosis for this mass?

A hypoechoic mass is imaged at the eight o'clock position, 4 cm from the left nipple. Using ultrasound guidance, a biopsy of this mass is done. Fibrocystic changes are reported histologically. What do you think? What would you recommend for this patient? The imaging and pathology findings are not congruent. Given the mammographic features of this mass, the likelihood of malignancy is high and a benign diagnosis on the cores is not acceptable.

In assessing this situation and problem solving in general, it is helpful to go back to basics. The pathologist should be asked if additional sectioning of the cores can be done, because sometimes the lesion is deep in the cores and it is possible that the lesion has not yet been examined histologically. If all available tissue has been sectioned and examined, the mammographic and ultrasound images should be reviewed. Do the images obtained during the biopsy confirm adequate positioning of the needle through the mass? Ideally, orthogonal images of the needle are obtained during the biopsy to document final needle positioning. With small lesions it is particularly important to document that the needle is associated with the mass longitudinally as well as in cross section (Fig. 4.4F [I and III] and Fig. 4.4G, I). There are times when the needle appears to be through the mass longitudinally, but the needle is actually just at the edge of the mass (and not in it) when the needle is imaged in cross section (Fig. 4.4F [II] and Fig. 4.4H). Lastly, you have to ask yourself: Does what is seen sonographically correlate with the mammographic finding?

In this patient, the pathologist has reviewed all available material and the images during the biopsy document adequate needle positioning. What do you think relative to the correlation of the lesion seen mammographically with what is imaged on ultrasound? At what clock position would you expect to find the mammographic finding, and at what distance should this lesion be from the nipple? The lesion is expected at the 8 o'clock position; however, in measuring back from the nipple, this lesion is closer to 8 cm and not 4 cm (Fig. 4.4J–L) from the nipple. When the patient is scanned at the 8 o'clock position, 8 cm from the nipple, a 9-mm spiculated mass with angular margins and associated shadowing is imaged at this site (Fig. 4.4M, N). This corresponds with the mass seen mammographically and its ultrasound features suggesting a malignant process correlate closely with those seen mammographically. Ultrasound guidance is used preoperatively to localize the mass at the 8 o'clock position, 8 cm from the left nipple (Fig. 4.4O). A 0.8-cm, well-differentiated invasive ductal carcinoma not otherwise specified is reported following the lumpectomy. No metastatic disease is identified in two excised sentinel lymph nodes [pT1b, pN0, pMX; Stage I].

Correlation is critically important. For clinical findings, does what is seen mammographically correlate with what the patient or the clinician is describing? In this situation, talking directly to the patient while doing a physical examination and the ultrasound study can provide the needed correlation. Specifically, ask the patient to show you the location of what she is feeling. For mammographic findings, is what is seen on the ultrasound the same thing as what is on the mammogram? If I am doing an ultrasound for a mammographic finding, I walk into the ultrasound room with an expected clock position in mind and an estimated distance from the nipple. This is my starting point for the physical examination and where I place the transducer. After I have evaluated this area fully, I scan the remainder of the quadrant or the breast as needed. For findings on magnetic resonance imaging (MRI), is what is seen on the ultrasound the same thing as what is on the MRI? Knowing slice thickness, the distance of the lesion with respect to the nipple can be estimated and used as the starting point for the ultrasound study. If an imaging-guided biopsy is done, are the imaging and histologic findings congruent? If they are not congruent, is the problem with the imaging or is there a possibility the lesion has not been evaluated histologically (i.e., has all tissue been sectioned?)?

Lastly, following excisional biopsies, is there correlation between clinical and imaging findings and the reported histology? Specimen radiography is used to confirm that a clinically occult, preoperatively (wire) localized lesion is excised and the location of the lesion in the specimen is marked for the pathologist, thereby assuring that the lesion is examined. However, for patients in whom preoperative localization and specimen radiography are not done, we have had situations in which a solid, water-density mass is reportedly excised based on palpable findings and a lipoma or other noncongruent lesion is reported histologically. In this situation either the lesion was not excised or the pathologist did not evaluate the lesion of interest in the specimen. In these patients, repeating the mammogram is helpful in determining whether the lesion of interest has been excised.

If we are methodical in our approach and provide the needed correlation at every step of the process, the clinical breast imager is a critical factor in optimizing patient care. We are in a unique position to provide the needed correlation among clinical, imaging, and histologic findings.

 

 

Patient 5

Figure 4.5. Diagnostic evaluation in 43-year-old patient presenting with a tender “lump” in the right subareolar area. Craniocaudal (A) and mediolateral oblique (B) views. Ultrasound image (C) of the right subareolar area at the site of the palpable finding. Photograph of the breast (D) and ultrasound image (E) of the right subareolar area, 72 hours following (C). Second patient presenting with a tender mass in the right subareolar area. On physical examination (F), there is erythema (long arrows) in the periareolar region laterally. Other pertinent observations include a periareolar scar (small arrows) and a healed fistula (double-headed arrow). This patient has had multiple recurrent episodes of subareolar abscess formation with prior surgery and fistula formation. Also note the horizontal inversion (arrowhead) of the nipple. It is postulated that this type of nipple inversion is a reflection of periductal fibrosis resulting from recurrent episodes of inflammation. In this patient, the ultrasound examination (G) demonstrates a lenticular-shaped complex cystic mass in the subcutaneous tissues of the subareolar area (as though dissecting through the subcutaneous tissues), a common ultrasound appearance of early subareolar abscess formation.

 

 

What are your observations?

There is increased density in the right subareolar area corresponding to the site of concern to the patient. In looking at the technical factors, the right breast is less compressible and a higher kilovoltage and milliamperage were needed for exposure when compared with the left breast. Physical examination and an ultrasound are indicated for further evaluation.

On physical examination, erythema and peau d'orange changes are noted in the periareolar area. Significant tenderness is elicited with gentle compression. A mass is palpated in the right subareolar area. Sonographically, a mass with posterior acoustic enhancement is imaged in the subareolar area, corresponding to the palpable finding (Fig. 4.5C). Based on the clinical presentation and the imaging findings, an ongoing inflammatory process is suspected. The patient is started on antibiotics.

The patient returns within 72 hours, describing progressive symptoms and purulent fluid draining from the periareolar area. On physical examination, erythema and peau d'orange changes are again noted, but these are now much more extensive and there is now a protuberant mass extending into the upper inner quadrant of the right breast. The overlying skin is thinned and there is a fistula draining purulent fluid at the areolar margin at the 1 o'clock position (Fig. 4.5D). On ultrasound, the abscess has increased in size (Fig. 4.5E). Given the rapid progression of symptoms, the formation of a fistula, and the subareolar location of this abscess, the patient is transferred to the hospital for consultation and surgical drainage.

When considering mastitis and abscess formation, what groups of patients should you consider?

Three different patient populations can be considered relative to mastitis and breast abscess formation. Most commonly, we associate these inflammatory conditions with women who are breastfeeding: This is puerperal mastitis. Reportedly, mastitis occurs in approximately 2.5% of women who breastfeed, and abscess formation affects >,1 in 15 of all women who breastfeed. In this patient population, Staphylococcus aureus is the most common causative agent. Patients with mastitis alone are usually treated effectively with antibiotics. If an abscess develops, percutaneous drainage can be helpful, and surgical drainage may be required for some patients. These patients are usually treated by their obstetricians and are not usually referred for imaging.

The second group of patients to consider in terms of breast infections is those with recurring subareolar abscess formation unrelated to nipple piercing or nipple rings. These patients are nonlactating, premenopausal women, most with a history of heavy smoking. As seen in this patient, many of these patients develop periareolar fistulas spontaneously (Zuska's disease). Squamous metaplasia involving the subareolar ducts is seen histologically in these patients, and it is postulated that this process leads to obstruction of the ducts, with inspissation of secretions, duct wall erosion, and the development of periductal mastitis and abscess formation. Antibiotic therapy alone is not usually effective in these patients. Although some have advocated percutaneous drainage, this also is not always effective, and wide surgical excision is required. Even in the patients who undergo surgical drainage, however, there is a high incidence of recurrence. With recurrent episodes, the nipple begins to flatten and some patients develop horizontal inversion centrally in the nipple (Fig. 4.5F, G). Bilateral abscess formation is seen in as many as a quarter of these patients, either simultaneously or at different times. A mixture of aerobic and anaerobic organisms is often cultured in these patients. It has been reported that these patients have a higher incidence of acne, hidradenitis suppurativa, and perineal inclusion cysts.

Lastly, peripheral mastitis or abscess formation can be seen unrelated to pregnancy and lactation. Rarely, some of these patients are diabetic; most of the women in this group, however, are otherwise healthy, with no identifiable source of infection. These patients respond well to antibiotic therapy and infection usually does not recur. They are also unlikely to present with bilateral findings.

 

 

Patient 6

Figure 4.6. Diagnostic evaluation, 54-year-old woman called back for a mass in the left breast detected on her screening study. Left mediolateral oblique (A), photographically coned view. Ultrasound images, radial (RAD) (B) and antiradial (ARAD) (C) projections corresponding to the area of the mammographic abnormality at the 7 o'clock position, zone 2 (Z2).

How would you describe the imaging findings?

A 2-cm, macrolobulated mass, with partially well circumscribed and indistinct margins, is seen mammographically. On ultrasound, a well-circumscribed, irregular mass with posterior acoustic enhancement is imaged at the 7 o'clock position, zone 2 (Z2), corresponding to the mammographic finding. Although a cyst is suspected, the presence of internal echoes and the irregular shape of this mass are such that this cannot be called a simple cyst. A cyst aspiration is undertaken. In discussing this with the patient, I tell her that if I do not obtain fluid, or if there is a residual abnormality postaspiration, I will do a needle biopsy and obtain tissue for histologic evaluation.

After establishing an approach that allows me to advance the needle parallel to the transducer, I clean the skin and use lidocaine to anesthetize the skin. Then, using ultrasound guidance, I inject lidocaine into the tissue leading up to, but taking care to not go into, the lesion. I use ultrasound guidance for administering the anesthesia and for doing the aspiration, even in those patients in whom the mass is palpable. Commonly, the advancing needle displaces the mass, or indents the wall, but does not penetrate into the mass (I think this explains many of the patients who present for evaluation of a palpable mass following attempted aspirations that yielded no fluid and yet we find a cyst corresponding to the palpable finding). By visualizing the trajectory of the advancing needle, I can gauge the amount of compression I need to apply to effectively immobilize the mass and the amount of controlled pressure I need to exert with the needle so that the cyst wall is punctured. Once the needle is in the mass, I pull the stylet out of the 20G spinal needle, attach a 10-mL syringe, and aspirate. I watch on real-time ultrasound as I aspirate to be sure there is no residual abnormality postaspiration. Also, in some patients, the needle may need to be redirected (i.e., the tip of the needle put against the cyst wall) during the aspiration to be sure that all of the fluid is aspirated. If I do not obtain fluid, I may try using an 18G spinal needle, and if I still do not obtain fluid, or if there is a residual abnormality postaspiration, I proceed with core biopsies using the 14G needle.

In this patient, 8 mL of greenish fluid is aspirated and no residual abnormality is seen following the aspiration. At this point, I inject 4 mL of air (50% of the aspirated fluid volume) into the cyst cavity, because it has been suggested that by doing this we can lower the incidence of cyst recurrence. The air does not hurt the patient, and if it is helpful in minimizing the likelihood of a recurrence, it can be beneficial to the patient. If I am concerned about the presence of a mural or intracystic abnormality, spot compression magnification views of the mass are done following the injection of air in the cyst (i.e., a pneumocystogram) to further evaluate the wall of the cyst.

As a routine, I discard aspirated fluid. Intracystic carcinomas are rare (0.5% of all carcinomas and >,0.1% of cysts), and even when an intracystic carcinoma is present, negative cytology is obtained in more than half of patients. I submit fluid for cytology if I obtain bloody fluid following an atraumatic tap, if there is a residual abnormality postaspiration, or if requested by the patient. It has also been recommended that fluid be submitted for cytology when a repeat aspiration is done in a patient who presents with rapid reaccumulation of fluid. As mentioned previously, in addition to submitting aspirated fluid for cytology, when a residual abnormality is seen postaspiration, I do core biopsies through the residual lesion.

Cysts have a variable mammographic appearance. They are usually round or oval masses with marginal characteristics that range from well circumscribed to obscured, to indistinct (particularly when inflamed). Mural and intracystic calcifications (milk of calcium) may be present. On ultrasound, simple cysts are well-circumscribed, anechoic masses with posterior acoustic enhancement and thin edge shadows. Less common appearances include the presence of intracystic echoes that during real time are characterized by movement (e.g., “gurgling”), and persistent, nonmovable echoes that sometimes have an S-shaped (yin-yang sign) interface with the more anechoic portion of the cyst. If the cyst is small and deep in the breast, posterior acoustic enhancement may not be apparent.

 

Patient 7

Figure 4.7. Diagnostic evaluation, 20-year-old patient presenting with a “lump” in the left breast. Ultrasound image (A) in the radial projection of the palpable finding at the 4 o'clock position, left subareolar area. Ultrasound images obtained during the ultrasound-guided aspiration. Preaspiration (B)image, documenting preaspiration needle positioning in the cyst, and postaspiration (C) image, demonstrating that there is no residual abnormality.

What would you do to evaluate this patient?

When would you obtain a mammogram for a 20-year-old woman?

Correlative physical examination and an ultrasound are our starting point in evaluating focal signs and symptoms in women under the age of 30 years or those who present during pregnancy or lactation regardless of age. A full mammogram is done only if breast cancer is suspected based on the clinical and ultrasound findings.

 

How would you describe the finding, and what is your differential?

What would you recommend next?

An oval, well-circumscribed complex cystic mass with posterior acoustic enhancement is imaged corresponding to the palpable finding. Although the appearance is somewhat atypical for a cyst, this is the main diagnostic consideration. Alternative possibilities include a fibroadenoma, papilloma, and pseudoangiomatous stromal hyperplasia. A galactocele, abscess, or posttraumatic or postsurgical fluid collection would be considerations in the appropriate clinical context. Although a cyst with atypical features is suspected, aspiration is undertaken.

The preaspiration image documents needle placement in the mass (Fig. 4.7B). A little less than 1 mL of serous fluid is aspirated. No residual abnormality is seen postaspiration (Fig. 4.7C). On ultrasound, simple cysts are described as anechoic, well-circumscribed masses with posterior acoustic enhancement and thin edge shadows. In some patients, however, cysts can be seen with internal echoes that shift in position as you image them in real time (i.e., gurgling cysts), persistent echoes that form an abrupt linear or S-shaped (yin-yang sign) interface with the cystic component of the mass, or high spicular echoes that do not shift in position. Gurgling cysts do not require aspiration unless the patient is symptomatic. Depending on your level of concern, relative to the latter two types of cysts, aspirations are not absolutely indicated.

 

Patient 8

Figure 4.8. Screening study, 47-year-old woman. Craniocaudal (A) and mediolateral oblique (B) views. No prior films are available. Ultrasound images in the radial (RAD) (C, D) projection of the mass in the left breast, at the 2 o'clock position, 3 cm from the left nipple. Screening study, 1 year following (A)and (B). Craniocaudal (E) and mediolateral oblique (F) views.

What do you think?

A round mass is present in the upper outer quadrant of the left breast. The patient is called back for additional evaluation, which includes spot compression views and an ultrasound. Differential considerations at this point include cyst, fibroadenoma (complex fibroadenoma, tubular adenoma), phyllodes, papilloma, pseudoangiomatous stromal hyperplasia, adenosis tumor, and focal fibrosis. Depending on the clinical context, sebaceous cyst, galactocele, postoperative or traumatic fluid collection, and an abscess are also in the differential. Given the seemingly circumscribed margins, malignancy is less likely; however, invasive ductal carcinoma not otherwise specified, medullary, mucinous, or papillary carcinoma, or metastatic disease, particularly if the patient has a known malignancy, are additional considerations.

Spot compression views (not shown) demonstrate a well-circumscribed mass with no associated calcifications. On ultrasound, a well-circumscribed, anechoic mass with posterior acoustic enhancement is imaged, consistent with a simple cyst. In an asymptomatic patient, this requires no additional intervention or short-interval follow-up. The patient is reassured that what we are seeing is not cancer, that cysts do not turn into cancer, and that they are common, with many women developing them at various times.

BI-RADS® category 2: benign finding. Annual screening mammography is recommended.

 

 

What would you do next?

The mass in the left breast has enlarged. In determining the next appropriate step, I review the evaluation done the previous year. I focus on the ultrasound study. If, as in this patient, the ultrasound demonstrates a “classic” simple cyst, I do not call the patient back for a repeat evaluation. Cysts fluctuate in size, and as long the patient remains asymptomatic, I do not call the patient back. If, in reviewing the prior ultrasound there is any question about the diagnosis of a cyst (e.g., internal echoes, shadowing, or irregular margins), I will call the patient back for a repeat ultrasound.

BI-RADS® category 2: benign finding. Annual screening mammography is recommended.

 

Patient 9

Figure 4.9. Diagnostic evaluation, 47-year-old patient presenting with a “lump” in the right breast. Craniocaudal (A) and mediolateral oblique (B) views, metallic BB used to mark location of palpable finding. Spot tangential (C) view of palpable finding. Ultrasound images in the radial (RAD) (D) and antiradial (ARAD) (E) of the palpable finding. Ultrasound image (F) after approximately 0.5 mL of grossly bloody fluid (G) is aspirated. Image of one of the core samples (H) demonstrating hemorrhagic tissue flanked by fatty tissue.

 

How would you describe the findings, and what would you recommend next?

A well-circumscribed mass is imaged corresponding to the palpable finding. On the tangential view, this appears to be a mixed-density lesion such that differential considerations include a lymph node, fat necrosis, hematoma, galactocele, or a fibroadenolipoma. The patient does not recall any recent trauma, and there is no history of a recent pregnancy.

 

How would you describe the findings, and what is your recommendation?

On physical examination a discrete, mobile, hard mass is palpated at the 10 o'clock position, 6 cm from the right nipple. A well-circumscribed, oval 1-cm mass that is nearly anechoic with posterior acoustic enhancement is imaged corresponding to the palpable finding. Although a cyst is suspected given the presence of internal echoes and following a discussion with the patient, an aspiration is undertaken.

The mass is atraumatically punctured using a 20G spinal needle and grossly bloody fluid is aspirated. Although several attempts are made to reposition the needle, no additional fluid is obtained and a residual abnormality persists. Core biopsies are done through the persistent abnormality. On inspection of the cores, a central area of hemorrhagic tissue (i.e., a lesion) is noted, on either side of which fatty tissue is seen. This correlates with what is seen mammographically: The lesion is surrounded by fat. The appearance of the cores, in conjunction with that of the aspirate, suggests either a hematoma or fat necrosis. The aspirated fluid is submitted for cytology and the cores are submitted for histologic evaluation. Predominantly blood and hemosiderin-laden macrophages are reported on the cytology. Fibroadipose tissue with granulation tissue, necrosis, hemosiderin-laden macrophages, chronic inflammation, fat necrosis, and foreign-body giant cells are reported on the cores. These findings are congruent with the clinical and imaging findings and the gross appearance of the cores. Annual mammography is recommended for this patient.

Patient 10

Figure 4.10. Diagnostic evaluation, 47-year-old patient presenting with a “lump” in the right breast. Craniocaudal (A) and mediolateral oblique (B) views with a metallic BB placed at site of the palpable finding, right breast. Spot compression view (C) of the palpable finding. Ultrasound images, radial (RAD) (D) and antiradial (ARAD) (E) projections of the palpable finding at the 6 o'clock position of the right breast, 2 cm from the nipple. Ultrasound images, radial (RAD) (D) and antiradial (ARAD) (E) projections of the palpable finding at the 6 o'clock position of the right breast, 2 cm from the nipple. Ultrasound images (F, G) obtained during aspiration. Ultrasound image (H), approximately 72 hours following the aspiration.

A round mass with partially well circumscribed margins is present in the right breast, corresponding to the site of clinical concern. The differential considerations for the mammographic findings in this patient include cyst, fibroadenoma (tubular adenoma, complex fibroadenoma), papillary lesion, focal fibrosis, pseudoangiomatous stromal hyperplasia (PASH), adenosis tumor, and phyllodes tumor. Depending on the clinical context, galactocele, postoperative or traumatic fluid collection, and an abscess are also in the differential. Given the seemingly circumscribed margins, malignancy is less likely; however, invasive ductal carcinoma not otherwise specified, medullary, mucinous, or papillary carcinoma, or metastatic disease, particularly if the patient has a known malignancy, are additional considerations. An ultrasound is indicated for further evaluation.

How would you describe the findings, and what is your recommendation?

There is no history of a recent pregnancy, trauma, surgery, or significant tenderness. On physical examination, a hard but mobile mass is palpated at the 6 o'clock position, 2 cm from the right nipple. A well-circumscribed mass with internal echoes and posterior acoustic enhancement is imaged corresponding to the palpable finding. Although a cyst is suspected, the echoes did not change in position during the ultrasound study and so an aspiration is undertaken. The mass is easily and atraumatically punctured, and approximately 8 mL of grossly bloody fluid is aspirated. No residual abnormality is seen postaspiration. The fluid is submitted for cytologic evaluation. Abundant red blood cells in a background of acellular debris are reported on the thin prep, and the cell block material is submitted for cytology. No malignant cells are identified.

The fluid reaccumulated in a matter of days. This, in combination with the bloody fluid obtained during the initial aspiration, suggests that further action is indicated. An excisional biopsy is performed, and an intermediate-grade, intracystic papillary carcinoma confined within a fibrous capsule (i.e., ductal carcinoma in situ) is reported on the excisional biopsy [pTis, pN0(sn), pMX; Stage 0]. There is no evidence of invasion. Atypical ductal hyperplasia is reported in the surrounding tissue.

Intracystic papillary carcinomas are rare, but they should be suspected if bloody fluid is aspirated following an atraumatic tap, if there is a residual abnormality following the aspiration, or if fluid reaccumulates rapidly. It is also important to emphasize that cytology is negative in a significant number (>.50%) of patients with intracystic carcinomas, so if there is a mural or intracystic component, core biopsies through these areas may be useful in establishing the correct diagnosis.

 

 

Patient 11

Figure 4.11. Diagnostic evaluation, 58-year-old patient who presents describing nipple discharge on the right. Craniocaudal (A) and 90-degree lateral (B)magnification views of the right breast following a ductogram. Magnification view, craniocaudal projection, right breast. Two intraductal lesions (arrows) are present (C) in a dilated duct. Excisional biopsy confirms the preoperative diagnosis of papillomas.

 

 

How should patients presenting with nipple discharge be evaluated?

What do you think of this ductogram?

For patients presenting with nipple discharge, a full mammogram is done (images not shown for this patient), history is obtained, and a physical examination is done. If it is determined that the nipple discharge is spontaneous, a ductogram is done. In this patient, the cannulated duct is dilated (arbitrarily, we use the cannula as an internal reference for duct size; normal ducts are one to three cannulas in diameter) and there are two lesions in the duct. The anterior-most lesion is a filling defect; the second lesion is obstructing a side branch of the involved duct (Fig. 4.11C, arrows). These are likely to be papillomas, but excisional biopsy is recommended. Excisional biopsy in this patient confirms the presence of intraductal papillomas with no atypia or other associated proliferative changes.

Ultrasound is sometimes used to evaluate women with nipple discharge. It is important to recognize that this is useful when the lesion or lesions are close to the nipple and in a dilated duct. However, this is not always the situation. Intraductal lesions can be found several centimeters away from the nipple in nondilated ducts and therefore are not always identified by ultrasound.

Ductography can provide information relative to the course of the abnormal duct, the number and location of lesions, and the likely etiology of the lesions. It is not appropriate to assume, as many surgeons do, that lesions causing nipple discharge are in the subareolar area or that the ducts containing the lesions are dilated and therefore identifiable intraoperatively. Even if the abnormal duct is identifiable at the time of surgery, the number and location of potential lesions cannot be established reliably through visual inspection intraoperatively. When patients are taken to the operating room because of discharge but in the absence of preoperative evaluation, blind excisions are done, with no assurance that the cause of the discharge has been excised and evaluated histologically. Following duct excision, the discharge is usually eliminated. The patient may be relieved, but what have we accomplished if the underlying lesion, possibly a ductal carcinoma in situ, has not been excised?

What information is useful in evaluating women who present with nipple discharge?

Obtaining a good history is a helpful starting point in evaluating patients who present describing nipple discharge. We ask the patient: “How did you notice the discharge?” Invariably, patients with significant nipple discharge provide one, or all, of three descriptions: They notice dark brown spots in the cup of their bra, dark spots on their night clothes, or they have just gotten out of a hot bath or shower, dried their breasts, and notice fluid coming from their nipple. This is spontaneous nipple discharge, and regardless of its appearance (e.g., clear, serous hem-occult-negative or bloody), requires further evaluation. It should be contrasted with expressed nipple discharge that is physiologic in etiology and does not usually require additional evaluation. A variable amount of fluid is found in normal ducts, so nipple discharge can be obtained from multiple duct openings, bilaterally, in most women following vigorous breast and nipple manipulation.

The next step in evaluating women with nipple discharge is physical examination. A bright source of light (e.g., a halogen lamp) focused on the nipple and magnification of the nipple are helpful. I start by examining the surface of the nipple for any crusting or a duct opening that appears erythematous or more patulous than the others. These may be indicators of a duct opening that needs to be cannulated for ductography. I next use an alcohol wipe to clean the surface of the nipple (this clears any keratin plugs that may be partially or completely occluding the duct opening) and I examine the breast for any palpable mass. In doing the exam, I check to see if I am able to elicit discharge. In many patients with an intraductal lesion, it is possible to identify the trigger point described by Haagensen. When you apply pressure over the trigger point, discharge is elicited. As you move away from the trigger point, the discharge stops. In many patients with an intraductal lesion, the discharge is projectile (shoots out at you and can hit you in the eyes if you are not careful) and copious.

If the patient provides a history of spontaneous nipple discharge, and single-duct discharge is elicited on physical examination, a ductogram is undertaken even in women with hem-occult-negative discharge. If fluid is difficult to obtain, and it originates from multiple duct openings bilaterally, ductography is not indicated. We do not routinely submit nipple discharge for cytology, and we do not routinely do hem-occult testing. A negative cytology report does not exclude significant pathology, however, and if atypical cells are described, the issue of establishing the presence and location of a lesion in the duct remains.

There is a misconception that only bloody nipple discharge is significant. On the contrary, clear or serous hem–occult-negative discharge may reflect the presence of underlying ductal carcinoma in situ, so if nipple discharge is spontaneous, it warrants additional evaluation. Although ductography is not a perfect test and it is associated with a 15% false negative rate, ductography can be helpful in identifying the presence of one or multiple intraductal lesions, the location and course of the duct containing the lesions, and the extent of the lesion. It seems to be a more helpful study than doing cytology and blind surgical excisions that can potentially cut through tumor or leave a lesion in the breast while eliminating the presenting symptom. Duct openings are closely apposed on the surface of the nipple, so a 15% false negative rate for ductography is not surprising. If a normal ductogram is obtained in a patient with a history and physical examination that is highly suggestive of an intraductal lesion, I assume the normal ductogram is a false negative study and ask the patient to return in 1 week for a repeat study.

How is a ductogram done?

The secreting duct opening is cannulated using a blunt-tipped 30G straight sialography needle. As a starting point, 0.2 mL of contrast is injected into the duct. The cannula is left in the duct and taped onto the breast. Craniocaudal and 90-degree lateral magnification views of the breast are obtained. The initial amount of contrast injected is small, so that the subareolar portion of the duct can be evaluated. If more contrast is injected at the onset, the density of the contrast may mask small lesions close to the nipple, resulting in a false negative study. By leaving the cannula in the duct, additional contrast material can be injected as needed.

 

What are the common causes of spontaneous nipple discharge, and what are the more common findings on ductography?

Papillomas, fibrocystic changes, duct ectasia, and breast cancer (usually ductal carcinoma in situ) are the more common causes of spontaneous nipple discharge. Findings on ductography include one or multiple filling defects, duct obstruction, wall irregularity, displacement of the duct, extravasation, and duct dilatation.

What can be done preoperatively to assure excision of the abnormal duct, and how can we facilitate the histologic evaluation of intraductal lesions?

In addition to doing diagnostic ductograms to establish the presence, location, and extent of intraductal lesion, we also do preoperative ductograms. A methylene blue contrast (1:1) combination is injected into the duct on the day of surgery. The contrast allows us to verify that we have cannulated the previously evaluated duct (i.e., the duct with the intraductal lesions), and the methylene blue stains the duct for the surgeon and pathologist. Having the ability to identify the duct stained in blue intraoperatively can effectively limit the excision to the abnormal duct. Of equal importance is facilitating identification of the abnormal duct for the pathologist, because even if the duct is dilated intraoperatively, it collapses as the fluid drains out after excision. The methylene blue is used to identify the duct grossly and direct the dissection for identification of the lesion. Rarely, if the intraductal lesion(s) is peripheral in location (i.e., not in the subareolar area) or if it is in a small branch of the main duct, the ductogram is done the day of surgery and used to guide a mammographically guided preoperative wire localization.

 

 

Patient 12

Figure 4.12. Diagnostic evaluation, 61-year-old patient presenting with nipple discharge on the left. Craniocaudal (A) and mediolateral oblique (B) views. Spot compression views of the left subareolar area, craniocaudal (C) and mediolateral oblique (D) views. Ductogram, magnification views of the subareolar area: 90-degree lateral (E), 90-degree lateral photographically coned (F), and craniocaudal (G) views, left breast. Magnification views of the subareolar area: 90-degree lateral photographically coned (H) and craniocaudal (I) views, left breast. Multiple filling defects are present in the opacified duct and there are several areas of wall irregularity (arrows). A portion of the cannula and multiple air bubbles are also evident on the craniocaudal view.

 

What do you think?

What would you do next?

A predominantly fatty pattern is present, with axillary lymph nodes noted bilaterally. Scattered round calcifications, arterial calcifications, and a 1-cm mass with relatively well circumscribed margins and a round calcification are noted on the subareolar spot compression views. Given the described nipple discharge, a more detailed history of the discharge and a physical examination are indicated.

The patient describes a several-month history of dark brown spots on her bra cup and dripping from her left nipple after hot showers. On physical examination, there is some crusting involving one of the duct openings on the left nipple. An alcohol wipe is used to clear the crusting. With gentle compression of the left subareolar area, discharge is elicited easily from the opening that originally had the crusting. This duct is cannulated and 0.2 mL of contrast is injected.

 

What do you think?

What is your differential, and what would you recommend?

The opacified duct is minimally distended. Multiple filling defects and areas of duct wall irregularity are present (arrows, Fig. 4.12H, I). The duct leads up to the mass noted in the spot compression views. On the craniocaudal view, round, well-defined lucencies consistent with air bubbles are seen in addition to the filling defects and wall irregularity. Differential considerations for the findings include papilloma(s), ductal carcinoma in situ (DCIS), or fibrocystic changes (e.g., hyperplasia, atypical ductal hyperplasia). Given the multiplicity of findings in conjunction with the wall irregularity, DCIS is a significant consideration in this patient. Excisional biopsy is recommended. A high-nuclear-grade ductal carcinoma in situ is diagnosed at the time of the excisional biopsy [pTis, pN0(sn) (i>2), pMX; Stage 0].

What are the likely causes of spontaneous nipple discharge, and what can be seen on ductography?

Papillomas are the most common cause of spontaneous discharge, diagnosed in slightly more than 50% of patients who present with spontaneous nipple discharge. One or multiple filling defects, duct obstruction, and wall irregularity are the most common findings on ductography in patients with papillomas. Most ducts that contain papillomas are dilated.

Fibrocystic changes represent the second most common cause of nipple discharge, reported in approximately 35% of patients. Filling of cysts with contrast, multiple filling defects, duct obstruction, and a more diffuse wall irregularity than that seen with a papilloma are the findings related to fibrocystic changes on a ductogram.

Distended ducts in close proximity to the nipple characterize duct ectasia; the opacified ducts assume a more normal caliber peripherally in the tissue. No focal intraductal abnormality is identified in patients with duct ectasia.

Breast cancer, commonly ductal carcinoma in situ, is diagnosed in 5% to 15% of patients who present with spontaneous nipple discharge. In some of these patients, a mass or pleomorphic calcifications may be seen on the mammogram; for many patients, however, the mammogram is normal. The findings on ductography overlap with those described for papillomas and include one or multiple filling defects, duct obstruction, wall irregularity, displacement of the duct, and contrast extravasation. In general, ducts involved with ductal carcinoma in situ are normal in caliber or only minimally distended. In contrast, ducts with associated papillomas are often moderately to significantly distended. Given the inability to distinguish between papillomas and ductal carcinoma in situ on ductography, excision is recommended for all patients identified with an intraductal lesion.

Potential pitfalls on ductography include false negative studies, air bubbles, masking of lesions, duct perforation, and contrast extravasation. The duct openings are closely apposed on the surface of the nipple. If too much discharge is elicited when trying to identify the duct opening, an adjacent opening may be flooded with fluid and inadvertently cannulated. If the history and physical examination are suggestive of an intraductal lesion, a normal ductogram is considered a false negative study and the patient is asked to return for a repeat ductogram. Because we do diagnostic and preoperative ductograms (repeat ductogram on the day of surgery using a methylene blue: contrast combination), we know that our false negative rate for ductography is approximately 15%.

In preparing for the ductogram, contrast is drawn into a 3-cc luer lock syringe and the contrast is run through the tubing of the sialography needle. Every effort should be made to eliminate air bubbles from the system prior to cannulation. If air bubbles inadvertently enter the duct system they are usually easy to distinguish from a lesion. Air bubbles are well defined, lucent, and shift in position between films. Intraductal lesions are characterized by irregular contours, and they do not shift in position between films.

A small amount (0.2 mL) of contrast is injected initially because otherwise, small lesions, particularly when close to the nipple, can be masked. Because we leave the cannula in the duct, we are able to inject additional contrast as needed to distend the more peripheral portions of the duct.

Perforation of a normal duct is rare; it takes significant pressure, and the patient describes a sharp pain when the duct is perforated and burning as soon as you inject contrast. Adhering to some simple guidelines during cannulation can prevent this possible complication. After identifying the secreting duct opening, I angle the cannula approximately 45 degrees, place the tip at the secreting duct opening, and straighten the cannula. It usually “falls” into the duct. If it does not, I gently twirl the cannula between my thumb and index fingers but do not apply any pressure or try to advance the cannula forcefully. If I have identified a trigger point, I try to angle the cannula in the direction of the trigger point. Sometimes, lifting the nipple up so as to “straighten” the duct in the subareolar portion can also be helpful. As I am manipulating the cannula, I repeatedly ask the patient if she is feeling anything sharp. If the patient describes any discomfort, I stop and reposition the cannula. Ductography is a painless procedure, so if the patient describes discomfort it is commonly an indication that the cannula is not in a duct opening.

Contrast extravasation can be seen in the context of duct perforation. In this situation, the patient describes a burning sensation as soon as you attempt to inject contrast, and an amorphous collection of contrast is seen in the subareolar area. Alternatively, peripheral extravasation can be seen as an amorphous collection of contrast surrounding the side branches of normal or hypoplastic ducts. In this situation, the patient experiences no discomfort as you start injecting the contrast, but she will describe a burning sensation after some volume of contrast has been introduced into the duct system. Opacification of lymphatic channels is sometimes also seen in patients with peripheral contrast extravasation.

 

 

Patient 13

Figure 4.13. Screening mammogram, 54-year-old woman. Craniocaudal (A) and mediolateral oblique (B) views. Craniocaudal (C) and mediolateral oblique (D) double spot compression magnification views of the left breast. Subtraction images (E, F) obtained from precontrast and sequential sagittal T1-weighted images done after the intravenous bolus administration of gadolinium. Ultrasound images, radial (RAD) (G) and antiradial (ARAD) (H)projections.

 

What do you think?

What is indicated next?

There is dense tissue. A cluster of calcifications is present posterolaterally in the left breast. Magnification views are indicated for further characterization.

BI-RADS® category 0: need additional imaging evaluation.

What is your differential and recommendation?

A cluster of pleomorphic calcifications is confirmed on the magnification views (Fig. 4.13C, D). Fibrocystic changes including hyperplasia, atypical ductal hyperplasia, columnar alteration with prominent apical snouts and secretions (CAPSS), and sclerosing adenosis as well as fibroadenoma, papilloma, and ductal carcinoma in situ (usually low- or intermediate-nuclear-grade) are in the differential for this cluster of calcifications. An imaging-guided biopsy is indicated.

BI-RADS category 4: suspicious abnormality, biopsy should be considered. A stereotactically guided biopsy is done, and ductal carcinoma in situ is reported on the cores.

What would you recommend next?

Surgical consultation is indicated. Additionally, we recommend magnetic resonance imaging (MRI) for all of our patients with a diagnosis of breast cancer. This is particularly helpful in women with dense tissue mammographically. The purpose of the MRI is to better evaluate the extent of disease, including the presence of multifocal or multicentric disease in the ipsilateral breast, and to further evaluate the contralateral breast for synchronous lesions.

Irregular and clumped linear enhancement is noted, corresponding to the area of the patient's known DCIS in the upper outer quadrant of the left breast (Fig. 4.13E). Additionally, two masses with kinetic curves demonstrating rapid wash-in and wash-out of contrast are present in the lower outer quadrant of the left breast (Fig. 4.13F). The patient is asked to return for ultrasound evaluation of the MRI findings.

A mass with associated shadowing is imaged at the 4:30 o'clock position of the left breast, 5 cm from the left nipple. An ultrasound-guided biopsy is done and an invasive ductal carcinoma is reported histologically. Given the presence of multicentric disease (synchronous lesions in different quadrants), a mastectomy is recommended. On the mastectomy specimen, two foci (1.5 cm and 1.2 cm) of intermediate-grade invasive ductal carcinoma and associated DCIS (cribriform and micropapillary) are identified, and metastatic disease is reported in one of two excised sentinel lymph nodes; an additional positive lymph node is reported following the axillary dissection [pT1c, pN1a, pMX; Stage IIA].

What are some of the current indications for magnetic resonance imaging (MRI) of the breasts?

The role of MRI in breast imaging is evolving, and appropriate indications vary, depending on availability and experience. Described indications include the evaluation of women with an identified malignancy, particularly if they have dense tissue mammographically. In these patients, unsuspected multifocal or multicentric disease, or a breast cancer in the contralateral breast, may be identified, potentially altering the management of the patient. Other described uses include the evaluation of patients with metastatic disease to the axilla with an unknown (but presumed) breast primary, to potentially identify the primary; in monitoring the response of a tumor in patients undergoing neoadjuvant therapy; for assessing the presence of residual tumor in patients with positive margins postlumpectomy; and in distinguishing tumor recurrence from scar tissue at a prior lumpectomy site. The use of MRI to screen women for breast cancer is indicated in patients with dense tissue mammographically and a high risk of breast cancer, particularly BRCA1 or -2–positive patients, or those at least 10 years after chest wall radiation for lymphoma.

Scanning protocols for magnetic resonance imaging of the breast are still evolving. Bilateral, simultaneous imaging of the breasts is critical in assessing areas of asymmetry, hormonal influence, or diffuse change. A dedicated breast coil, at least a 1.5-T magnet, 1-mm spatial resolution in all planes, high temporal resolution, a 2-mm slice thickness, subtraction or fat suppression, and biopsy capability are minimal requirements for doing magnetic resonance imaging of the breast. As with mammography and ultrasound, meticulous technique is critical. This includes positioning of the breasts in the coil with no significant compression as well as arm positioning, and which vein is used to administer the contrast bolus for the dynamic portion of the scan. Lesions are assessed for signal characteristics and morphology, using T1- and T2-weighted images and for contrast enhancement patterns and kinetic analysis on the dynamic T1 scans obtained following the administration of contrast.

Lesions with high T2 signal intensity are usually benign and include cysts, lymph nodes, and some myxomatous fibroadenomas. Except for some mucinous carcinomas and necrotic tumors, most malignant lesions are characterized by low T2 signals. The shape of masses detected on MRI can be described as round, oval, lobulated, or irregular, and their margins as smooth, irregular, or spiculated. Margin analysis is best done on the first post-contrast image and is dependent on the size of the lesion and spatial resolution. In considering contrast enhancement, lesions may demonstrate homogenous or heterogeneous enhancement, central or rim enhancement, or enhancing internal septations. Alternatively, dark internal septations may be seen in a mass, reflecting the lack of enhancement. Homogeneous enhancement is more common in benign lesions. Inflammatory cysts and fat necrosis may exhibit rim enhancement. Inflammatory cysts, however, are bright on T2-weighted images, and the use of fat-suppression images, in conjunction with mammographic findings and clinical history, are helpful in the diagnosis of fat necrosis.

Enhancement can also be seen without the presence of a mass. In these situations, descriptive terms in the ACR lexicon for lesions detected on MRI include: focal area, linear, ductal (i.e., pointing toward the nipple and possibly branching), regional (not conforming to expected distribution of a duct), segmental (triangular enhancement with the apex toward the nipple), multiple regions, or diffuse. Non-mass-like enhancement can be further qualified as homogeneous, heterogeneous, stippled/punctate, clumped, reticular/dendritic, symmetric, or asymmetric. Linear enhancement, particularly when clumped or irregular, is suggestive of DCIS.

With contrast administration and dynamic imaging, kinetic data can be evaluated. Breast cancers typically enhance rapidly (i.e., “wash in”), with the enhancement stabilizing (plateau) or gradually decreasing in signal intensity (i.e., “wash out”). This is thought to reflect tumor neovascularity and shunting. Normal tissue, and most benign lesions, demonstrate gradual and continuous contrast enhancement. In describing the signal intensity–time curve, slow, medium, and rapid are used to describe the enhancement pattern of a mass within the first 2 minutes or when the curve starts to change. The delayed phase of the curve (after the first 2 minutes or after the curve starts to change) should be described as persistent if the enhancement continues to increase over time, plateau if the signal intensity does not change over time after the initial increase, or wash-out if the signal intensity decreases after the initial rise.

 

 

Patient 14

Figure 4.14. Screening mammograms, 50-year-old woman, craniocaudal (A) and mediolateral oblique (B) views, left breast, photographically coned to a mass in the upper outer quadrant of the left breast. Craniocaudal (C) and mediolateral oblique (D) views, 1 year before (A) and (B), left breast, photographically coned to the mass in the upper outer quadrant of the left breast.

How would you describe the findings, and why is there a metallic clip in the breast?

An oval, well-circumscribed mass is present in the upper outer quadrant of the left breast. A metallic clip is present in the mass, consistent with a prior imaging-guided core biopsy. Metallic clips are deployed at the time of imaging-guided biopsy procedures when complete removal of the lesion may occur as a result of the biopsy (e.g., biopsy of a cluster of calcifications or small mass using an 11G vacuum-assisted device). If the lesion is diagnosed as a malignancy or high-risk lesion, and it is removed in its entirety, the clip marks the location of the original lesion so that the tissue around the lesion can be localized and evaluated histologically for residual tumor at the time of the lumpectomy. In reviewing the prior biopsy report in this patient, a fibroadenoma is reported. The mass is now larger.

What is your recommendation at this point?

Because the mass is enlarging, is an excisional biopsy indicated?

The original diagnosis of a fibroadenoma is congruent for a well-circumscribed oval mass seen mammographically. In premenopausal women, fibroadenomas can enlarge, and a change in size alone does not constitute an indication for excisional biopsy. It has been reported that volume growth rates of >,16% per month in women under the age of 50 years, and 13% per month in those over the age of 50 years, or up to a 20% mean change in dimension in a 6-month interval, regardless of age, are acceptable, and the patient can be followed.

In this patient, the mass has enlarged, but the change in size falls within the acceptable limit. Nevertheless, the next step I take is to call pathology and request a review of the previous biopsy material in the context of a mass that has increased slightly in size. The pathologist confirms the diagnosis of a fibroadenoma; specifically, that the histologic findings do not suggest the possibility of a phyllodes tumor. Annual mammography is recommended for this patient.

BI-RADS® category 2: benign finding.

In most patients, the diagnosis of a fibroadenoma on core biopsy is reliable. In some women, the distinction between fibroadenoma and phyllodes tumor is an issue. In these patients, appropriate management decisions require placing the imaging findings in the proper clinical and pathologic context. Fibroadenomas are common lesions in younger, premenopausal women. Phyllodes tumors are uncommon lesions that occur predominantly in perimenopausal or postmenopausal women.

Fibroadenomas and phyllodes tumors are biphasic (fibroepithelial) lesions, arising in the lobules and characterized by proliferating epithelial and stromal elements. It is the cellularity of the stroma, not the appearance of the epithelial elements, that is used to distinguish between these lesions. In young women, fibroadenomas may be characterized as having a cellular stroma. As estrogen levels decrease with advancing age, epithelial elements and stromal cellularity in fibroadenomas normally decrease and hyalinization (i.e., fibrosis) increases. When fibroadenomas are described as “cellular,” or when the descriptive term “hypercellular stroma” is used relative to a fibroadenoma, particularly if it is an older, perimenopausal or postmenopausal patient, a direct discussion with the pathologist regarding the possibility of phyllodes tumor is helpful. If the distinction cannot be made reliably based on the core samples, excisional biopsy is recommended.

 

Patient 15

Figure 4.15. Screening study, 62-year-old woman. Craniocaudal (A) and mediolateral oblique (B) views, left breast. Craniocaudal (C) and mediolateral oblique (D) views, left breast, 1 year prior to images shown in (A) and (B). Craniocaudal (E) and mediolateral (F) views, left breast, 2 years prior to images shown in (A) and (B). Ultrasound images, radial (RAD) (G) and antiradial (ARAD) (H), 1 year ago, at the time of (C) and (D).

What do you think?

What would you do next?

Dense glandular tissue with several coarse calcifications, possibly reflecting fibroadenomas undergoing hyalinization and calcification, is noted. Additionally, there is a round, well-circumscribed mass in the lower outer quadrant of the left breast. Prior films would be helpful in determining the next appropriate step.

 

 

What do you think?

The solid, well-circumscribed mass in the left breast is enlarging. A fibroadenoma was reported on core samples obtained following an ultrasound-guided biopsy done 1 year ago. There is no history of hormone replacement therapy. What do you think? Are the clinical, imaging, and histologic findings congruent? Given the most recent mammogram (Fig. 4.15A, B), what is your main concern, and what would you recommend at this time?

The diagnosis of a fibroadenoma is a congruent diagnosis for a round, well-circumscribed solid mass, particularly in younger, premenopausal patients. In such a patient, an increase in the size of a fibroadenoma diagnosed with a needle biopsy is not an absolute indication for excision unless the change in size is significant. In a 61-year-old woman, however, particularly if she is not on hormone replacement therapy, the diagnosis of a fibroadenoma should be considered carefully and discussed with the pathologist directly. Fibroadenomas may develop in the early postmenopausal period, particularly if a patient is started on hormone replacement therapy. They are not expected to develop and enlarge years following menopause in a patient with no history of hormone replacement therapy. The pathologist should be asked specifically about the possibility of a phyllodes tumor. In this postmenopausal patient, the original diagnosis should have been challenged and now, a year later, as you review all available studies and note the progressive change in the size of this mass, an excisional biopsy is indicated.

A benign phyllodes tumor is diagnosed on the excised tissue.

What are phyllodes tumors, and how do they present?

Phyllodes tumors are rare, representing between 0.3% and 1% of all breast tumors. They are biphasic (fibroepithelial) tumors, diagnosed more commonly in perimenopausal and postmenopausal patients. The median age (45 years) of patients with phyllodes tumors is approximately 15 years higher than the median age of women who present with fibroadenomas. Phyllodes tumors commonly present as a single, well-circumscribed, hard mass; rarely, patients may present with multiple phyllodes tumors in one or both breasts. These tumors develop from lobules, and their resemblance to fibroadenomas has led some to suggest that they arise from pre-existing fibroadenomas. Alternatively, some have postulated that they arise de novo.

Histologically, phyllodes tumors are characterized by the presence of clefts or cystic spaces lined by epithelial cells and a cellular stroma. The epithelial elements in these lesions are normal and similar to those seen in fibroadenomas. It is the appearance of the stroma that is used to distinguish phyllodes tumors from fibroadenomas. Attempts have been made to subclassify phyllodes tumors into malignant, benign, and borderline lesions based on their margins, stromal cellularity and overgrowth, stromal cell atypia, and mitotic activity. Features that are suggestive of malignancy include the presence of infiltrative margins, marked stromal cellular overgrowth, moderate to marked atypia of the stromal cells, and 10 or more mitotic figures per 10 high-power fields. Features of benign tumors include expansile margins, moderate stromal cellularity, minimal atypia of the stromal cells, and 0 to 4 mitoses per 10 high-power fields. A borderline tumor is described when a lesion demonstrates expansile or infiltrative margins, moderate atypia of the stromal cells, and 5 to 9 mitoses per 10 high-power fields. Rarely, sarcomatous elements including angiosarcoma, liposarcoma, chondrosarcoma, myosarcoma, or osteosarcoma are described in the stroma of phyllodes tumors.

It is important to emphasize that fibroadenomas in younger women are characterized by the presence of epithelial elements and a stroma that may be described as cellular. As patients age, and estrogen levels decrease, the epithelial elements and cellularity of the stroma decrease, and fibrosis and hyalinization of the lesion occurs. The diagnosis of a “cellular” fibroadenoma is acceptable in young women (teens and 20s, early 30s), but care should be exercised in accepting the diagnosis of a “cellular” fibroadenoma, or one characterized as having a “hypercellular stroma,” in perimenopausal and postmenopausal women. In this latter group of patients, the pathologist needs to be asked specifically about the possibility of a phyllodes tumor, and if this is a concern, excisional biopsy is appropriate.

Local recurrence is the main concern in patients diagnosed with a phyllodes tumor, but distant metastases and death can occur. Wide surgical excision is critical in minimizing the likelihood of local recurrence. A hematogenous route of spread is described in patients with metastatic disease. Consequently, axillary dissections are not indicated for patients with phyllodes tumors. Benign phyllodes tumors do not usually metastasize, have a lower incidence of local recurrence, and the interval to recurrence is longer.

Patient 16

Figure 4.16. Screening study, 64-year-old woman. Left craniocaudal (A) and mediolateral oblique (B) views, photographically coned. Ninety-degree lateral (C) view documenting final wire positioning following an ultrasound-guided wire localization of both lesions [arrows in (D)] in the upper outer quadrant of the left breast.

How would you describe the findings?

Two masses are present in the upper outer quadrant of the left breast. The more superior and lateral of the lesions is characterized by microlobulated and irregular margins. The second mass has partially well circumscribed margins. Compared with prior studies (not shown), these findings represent a change. Imaging-guided biopsy is done of the more superior and lateral of the lesions. A sclerosing lesion with associated atypical ductal hyperplasia and mucin dissecting in the sclerosing lesion and adjacent stroma is reported on the cores.

Is this congruent with the imaging findings?

At this point, what is your recommendation for this patient and why?

The mammographic and histologic findings are congruent. However, with a sclerosing lesion, atypical ductal hyperplasia, and dissecting mucin reported in the lesion and stroma, an excisional biopsy is recommended. Both masses, seen mammographically, are excised following wire localization (Fig. 4.16C, D). An intraductal papilloma with apocrine atypical ductal hyperplasia and adjacent mucocele-like tumor were reported for the lesion biopsied previously. An intraductal papilloma with no atypia is reported for the second excised mass.

What are mucocele-like lesions of the breast, and how should they be managed following diagnosis on core biopsy?

Mucocele-like lesions of the breast are made up of multiple cysts or dilated ducts containing mucinous material that is extruded into the surrounding stroma. The epithelial cells lining these cysts or ducts are uniformly flat or cuboidal to columnar in appearance. Although they were initially described as benign lesions, there are now reports in the literature of associated atypical ductal hyperplasia, ductal carcinoma in situ (DCIS), or invasive carcinoma with some of these lesions. The associated DCIS is usually micropapillary or cribriform type, and the invasive lesion is usually mucinous carcinoma. Consequently, the diagnosis of a mucocele-like lesion, or the presence of mucinous material dissecting in the stroma on core biopsy samples, should prompt consideration of an excisional biopsy of the lesion.

What imaging findings have been reported for mucocele-like lesions of the breast?

In screening programs, these lesions are usually diagnosed on core biopsies done for indeterminate or suspicious microcalcifications; some of the described calcifications are coarse and eggshell-shaped. Masses with margins ranging from well circumscribed to indistinct, with or without associated calcifications, have also been reported. On ultrasound, cysts with noncalcified or calcified mural nodules, hypoechoic masses characterized by low-level internal echoes, or tubular structures with low-level internal echoes may be seen in these patients.

 

Patient 17

Figure 4.17. Screening study, 42-year-old woman. Craniocaudal (A) and mediolateral oblique (B) views. Craniocaudal (C) and mediolateral oblique (D)spot compression views, left breast. Ultrasound images, radial (RAD) (E) and antiradial (ARAD) (F) projections corresponding to the area of mammographic concern, 6 o'clock position, 2 cm from the left nipple.

What do you think?

A potential mass is described in the left breast. If no prior films are available, or if this finding represents an interval change, the patient should be called back for spot compression views and possibly an ultrasound.

BI-RADS® category 0: need additional imaging evaluation.

What do you think now, and what would you recommend next?

At what clock position would you place the ultrasound transducer to find this mass?

A mass is confirmed on the spot compression views. Although the margins are obscured on the craniocaudal projection, they are better seen and appear partially well circumscribed on the oblique projection. Benign considerations in a 42-year-old woman include cyst, fibroadenoma (complex fibroadenoma, tubular adenoma), papilloma, focal fibrosis, pseudoangiomatous stromal hyperplasia, papilloma, adenosis tumor, phyllodes tumor, and granular cell tumor. Depending on history and clinical findings, an inflammatory process, posttraumatic/postsurgical fluid collection, and a galactocele might also be included in the differential. Malignant considerations include invasive ductal not otherwise specified, medullary carcinoma, or metastatic disease. Although they are less likely given the patient's age, mucinous and papillary carcinomas are also in the differential.

 

 

What do you think based on the ultrasound images, and what is your recommendation?

On ultrasound, a hypoechoic oval mass is imaged at the 6 o'clock position, 2 cm from the left nipple, corresponding to the mammographic finding. The margins are not well defined and there is no posterior acoustic enhancement; although you may be tempted to call this a cyst, it does not fulfill the diagnostic criteria for a cyst and therefore further evaluation is indicated.

How would you approach this patient's evaluation?

For patients in whom a cyst is a possibility, I approach their interventional procedures in a stepwise manner. The first step, after injecting lidocaine in the skin and expected needle course, is to attempt an aspiration. If no fluid is obtained, or a residual abnormality is noted following aspiration, I do core biopsies through the mass. In this patient, no fluid is aspirated and atypical ductal hyperplasia (ADH), apocrine type, is reported on the core biopsies.

Is this histology concordant with the imaging findings?

What would you recommend next, and why?

Pleomorphic calcifications with round, punctate, and amorphous forms are the most common mammographic finding reflecting the presence of atypical ductal hyperplasia (ADH) and ductal carcinoma in situ (DCIS). Rarely, ADH and DCIS can present as parenchymal asymmetry, distortion, or a mass with well-circumscribed, often macrolobulated margins that may be further characterized as indistinct and sometimes spiculated. So the diagnosis of ADH in this patient with a mass is concordant. However, depending on whether a 14G automated spring-loaded or an 11G vacuum-assisted device is used for sampling, ADH is upgraded to DCIS or invasive ductal carcinoma on excision in as many as 56% and 27% of patients, respectively. Consequently, excisional biopsy is the appropriate management of patients diagnosed with ADH on core biopsy, regardless of the finding (e.g., calcifications, mass, or distortion).

In this patient, an excisional biopsy is done following an ultrasound-guided preoperative wire localization of the mass in the left breast. A 1.1-cm ductal carcinoma in situ, apocrine type, is diagnosed in the excised tissue [pTis, pNX, PMX; Stage 0].

 

Patient 18

Figure 4.18. Diagnostic evaluation, 72-year-old patient called back for evaluation of calcifications detected in the left breast on her screening mammogram. Craniocaudal (A) and mediolateral oblique (B) double spot compression magnification views, left breast. Specimen radiograph (C) of excised tissue.

How would you describe the findings?

A cluster of round, punctate, and amorphous calcifications is confirmed on the double spot compression magnification views. No linear forms, linear orientation, or change in configuration of the calcifications (i.e., no milk of calcium) is noted on the magnification views. Fibrocystic changes including hyperplasia, atypical ductal hyperplasia, columnar alteration with prominent apical snouts and secretions (CAPSS), and sclerosing adenosis, as well as fibroadenoma, papilloma, and ductal carcinoma in situ (usually low- or intermediate-nuclear-grade) are in the differential for this cluster of calcifications. Biopsy is indicated.

BI-RADS® 4: suspicious abnormality, biopsy should be considered.

Stereotactically guided core biopsies are done on this patient. Depending on the size of the lesion, the needle used (e.g., 14G, 11G, 8G), the number of cores taken, and the device used for the biopsy (e.g., vacuum-assisted or wire basket), the lesion may actually be completely “excised” during the biopsy. If this is a possibility, a metallic clip needs to be deployed at the time of the biopsy so that the location of the original lesion is marked. If a malignancy or a high-risk lesion is diagnosed, the clip deployed at the time of the imaging-guided biopsy can be used to localize the area for excision and further histological evaluation. Craniocaudal and 90-degree lateral views are obtained after a clip is deployed to document the location of the clip immediately following the biopsy. In this patient, a 14G needle in a spring-loaded device is used for the biopsy. Complete removal of the lesion is not expected, so no clip is deployed. Atypical ductal hyperplasia with associated microcalcifications arising in a background of columnar alteration with prominent apical snouts and secretions (CAPSS) is reported on the core biopsy.

What is indicated next?

A diagnosis of atypical ductal hyperplasia on core biopsy requires excisional biopsy. This is done following preoperative wire localization of the calcifications in the left breast.

The calcifications are confirmed to be in the specimen (Fig. 4.18C). Based on this radiograph, the location of the calcifications is marked for the pathologist. Residual atypical ductal hyperplasia arising in a background of columnar alteration with prominent apical snouts and secretions (CAPSS) with associated microcalcifications is reported on the excised tissue. The pathologist comments that “The area of CAPSS is partially involved with a monotonous low-grade cell population which is insufficient in extent for a diagnosis of low-grade ductal carcinoma in situ.” Atypical ductal hyperplasia (ADH) is considered a high-risk marker lesion. Patients with ADH have four to five times increased risk for developing breast cancer compared to the reference population. This risk is increased further in women with ADH and a family history of breast cancer.

A contiguous layer of epithelial cells and an intermittent basilar layer of myoepithelial cells line the basement membrane of normal ducts and lobules in the breast. Hyperplasia refers to an increase in the number of cells (usually epithelial, though in some processes it is the myoepithelial cells that are hyperplastic) lining the ducts. In usual hyperplasia, the number of epithelial cells lining the ducts is increased and secondary spaces in the ducts are irregular in size and shape and commonly elongated or slitlike. In atypical ductal hyperplasia, the number of cells lining the ducts is increased, and secondary spaces of varying sizes and shapes, though somewhat more rigid and fixed than those seen in usual hyperplasia, are present in the duct. These proliferative changes in the duct can progress to ductal carcinoma in situ (usually low- or intermediate-nuclear-grade), characterized by a monomorphic cell population and rigid secondary spaces in the duct. Descriptive terms used for these types of DCIS include cribriform, micropapillary, and solid. This spectrum of cellular changes in the duct is not typically associated with rapid cell proliferation or central necrosis. The calcifications that develop in association with these proliferative processes develop in secretions (not in necrotic debris). The resulting calcifications are typically variable in density (possibly reflecting the variably sized spaces in which they develop) and pleomorphic, including round, punctate, and in some patients amorphous (tiny calcifications below the resolution obtainable with magnification views in a patient).

These processes involve the duct in a multifocal manner so that in some patients you can have hyperplasia next to DCIS, next to another area of hyperplasia, next to ADH, etc. Several studies have reported the presence of ADH in a peripheral location to areas of DCIS. In some patients, the proliferative changes have no associated calcifications. In other patients, the proliferative changes have associated calcifications but these are not necessarily closely associated with the malignant cells. Consequently, when you target these calcifications you may or may not be targeting the malignant cells. Additionally, the distinction between ADH and DCIS can be problematic, particularly given the small amounts of tissue submitted following imaging-guided needle biopsies. There is a need to increase the number and size (e.g., 14G vs. 11G) of the cores and a need to recommend excision if ADH is reported following core biopsies.

The complexity of the situation in managing some patients with these processes is further compounded because, although efforts have been made to define and standardize classification schemes for these processes, the diagnosis of hyperplasia, atypical ductal hyperplasia, and low-nuclear-grade DCIS remains subjective. Also, as described in this patient, some of the criteria for DCIS may be present, but the changes are insufficient to qualify for the diagnosis. A study by Rosai in 1991 reported that there was no agreement on a diagnosis in 17 borderline cases submitted for review to five leading breast pathologists, and the diagnoses rendered in some of the cases ranged from hyperplasia to DCIS.

This spectrum of disease should be contrasted with DCIS that is characterized by rapid cell proliferation (high thymidine labeling) and central necrosis. These lesions are thought to arise de novo in the duct, without antecedent hyperplasia or atypical hyperplasia. Although these are usually high-nuclear-grade cells, some may be intermediate- or low-nuclear-grade. The cells lining these ducts are pleomorphic, do not demonstrate polarization relative to the duct lumen, and have multiple nucleoli. Mitotic figures, cell necrosis, and autophagocytosis may be seen involving the cells lining the ducts. In these patients, the malignant cells circumferentially narrow the duct lumen and there is necrotic debris in the center of the duct. The calcifications seen mammographically develop in necrotic debris and consequently are closely apposed to the malignant cells (i.e., the calcifications are molded by the proliferating cells). In targeting this type of proliferative process, when you target the calcifications you target the malignant cells. The number and size of the cores is not as critical in these patients. If you remove calcifications in your cores, you are likely to have made the diagnosis of DCIS.

 

Patient 19

Figure 4.19. Screening study, 69-year-old woman. Right craniocaudal (A) and mediolateral oblique (B) views. Craniocaudal (C) and mediolateral oblique (D), double spot compression magnification views, right breast.

How would you describe the findings?

Several coarse, dystrophic-type calcifications are present in the right breast. Additionally, a cluster of calcifications, some of which may be linear with possible linear orientation, is noted in the upper outer quadrant of the breast anteriorly. Magnification views are indicated for further characterization.

BI-RADS® category 0: need additional imaging evaluation.

 

 

The double spot compression magnification views confirm the presence of linear calcifications demonstrating a linear orientation. The margins of these calcifications are irregular and there are associated clefts in the calcifications. Ductal carcinoma in situ (DCIS) with associated central necrosis (usually high-nuclear-grade) is the primary consideration with calcifications having these features. Biopsy is indicated.

What is your recommendation?

BI-RADS® category 4: suspicious abnormality, biopsy should be considered.

An invasive mammary carcinoma with predominantly lobular features and ductal carcinoma in situ, high-nuclear-grade, with central necrosis is diagnosed on the core samples. A 1.2-cm, grade III, invasive ductal carcinoma with associated high-nuclear-grade ductal carcinoma in situ with central necrosis is diagnosed on the lumpectomy specimen. The sentinel lymph node is negative for metastatic disease [pT1c, pN0, pMX; Stage I].

Ductal carcinoma in situ used to be considered a “rare” disease, with only scattered descriptions in the literature regarding its histologic appearance and biologic significance. With the widespread use of screening mammography, and our ability to detect and biopsy microcalcifications, DCIS now constitutes a significant proportion of the breast cancer that is diagnosed and treated. Driven by mammographic findings, our knowledge and understanding of this disease process has been significantly advanced in the last two decades. It is now recognized that DCIS is not one disease but several diseases characterized by clinical, mammographic, and biologic heterogeneity. Based on histology, biologic markers, and associated invasive lesions, we can consider at least two main paths of origins. One group of DCIS arises or evolves through proliferative changes in the duct that include hyperplasia, atypical ductal hyperplasia, and ductal carcinoma in situ. These proliferative lesions coexist and are multifocal in the involved duct. They are characterized by low rates of proliferation, long intraductal phases, and not all of the DCIS arising through this pathway is thought to progress to invasion. In some patients, these are thought to be precursors for low- or intermediate-grade invasive ductal carcinomas.

In contrast, a second group of DCIS develops in the duct without progressing through hyperplasia and atypical hyperplasia. This type of DCIS is characterized by rapid cell proliferation with high thymidine labeling rates, central necrosis in the duct, and a short intraductal phase. These types of lesions are all thought to progress to invasive disease (i.e., obligate invaders), giving rise to poorly differentiated invasive ductal carcinomas. Low-nuclear-grade DCIS is not thought to evolve (i.e., is not a precursor) to high-nuclear-grade DCIS.

Pleomorphic calcifications are the most common mammographic finding associated with DCIS. Depending on the underlying process in the duct, the calcifications have a variable appearance. In the first group of proliferative processes, the calcifications have been described as developing in secretions. Although they are pleomorphic and demonstrate variation in density, they are usually round, punctate, or amorphous. Multiple clusters may be seen. These calcifications may be associated with hyperplasia, atypical ductal hyperplasia, and DCIS (usually a low- or intermediate-grade DCIS). It is also important to recognize that these processes may be present in the breast without any associated calcifications. With this type of proliferative process, we underestimate the extent of disease in nearly 50% of patients. When doing biopsies of this type of calcification, it is important to assure adequate sampling by either increasing the number and size of the cores or even considering excisional biopsy in some patients. Removing some of the calcifications in these patients does not assure a correct diagnosis, as reflected by the high percentage of ADH that is upgraded to DCIS or invasive cancer when ADH is excised following diagnosis on core biopsy.

When there is central necrosis in the duct reflecting rapidly proliferating cells, the calcifications are often linear and may demonstrate a linear orientation. In this type of proliferative process, the calcifications are intimately associated with the malignant cells; the calcifications are being molded by the proliferating cells and develop in the necrotic debris. Targeting the calcification in essence targets the malignant cells. If calcifications are removed in one or two cores, you are likely to have made the diagnosis. Complete workups with optimal magnification views are helpful in these patients, because mammography is good at estimating the extent of the disease. Disease is found where we see the calcifications. In patients with lesions occupying several centimeters, the use of multiple wires for the preoperative localization facilitates complete removal of the lesion. Aggressive pursuit of these types of calcifications is critical because of the short intraductal phase of the disease and the almost certain, and in some patients rapid, development of invasive disease (often poorly differentiated).

 

Patient 20

Figure 4.20. Diagnostic evaluation, 51-year-old patient called back for calcifications detected in her right breast on the screening study. Double spot compression magnification views, craniocaudal (A) and mediolateral oblique (B) projections.

 

How would you describe the findings?

The magnification views confirm the presence of two adjacent clusters of calcifications. The calcifications composing the clusters demonstrate pleomorphism and variable density; however, there are no linear forms and there is no linear orientation. The calcifications do not change significantly in configuration between the craniocaudal and oblique projections (i.e., these do not reflect milk of calcium). Although there are well-defined round and oval calcifications, some of these would fall under the “amorphous” terminology currently provided by the ACR lexicon. It is important to recognize that these are not really amorphous but rather tight clustering of punctate calcifications that fall below the resolution of the magnification we can obtain when imaging a patient. If these are magnified three or four times, as can be done with a specimen, some of the seemingly amorphous calcifications can be resolved into individual, tightly clustered punctate calcifications. Fibrocystic changes including hyperplasia, atypical ductal hyperplasia, columnar alteration with prominent apical snouts and secretions (CAPSS), and sclerosing adenosis, as well as fibroadenoma, papilloma, and ductal carcinoma in situ (usually low- or intermediate-nuclear grade, with no central necrosis) are in the differential for these clusters of calcifications. A biopsy is indicated.

BI-RADS® category 4: suspicious abnormality, biopsy should be considered.

A stereotactically guided core biopsy is done, and fibrocystic changes including sclerosing adenosis, CAPSS with associated atypia, and atypical lobular hyperplasia are reported on the core samples. In discussing the findings directly with the pathologist, he confirms that the calcifications are found in sclerosing adenosis and CAPSS; the atypical lobular neoplasia is noted incidentally in surrounding breast tissue, with no associated calcifications.

At this point, what do you recommend for this patient?

Given the presence of CAPSS with associated atypia, and incidentally identified atypical lobular neoplasia, excisional biopsy is recommended for this patient. No malignancy is identified on the excisional biopsy, and the patient is returned to annual screening.

Lobular neoplasia is a term used to describe a spectrum of proliferative changes in the acini of lobules that ranges from atypical lobular hyperplasia (ALH) to lobular carcinoma in situ (LCIS). Continuous with that of the ducts, a two-cell layer above the basement membrane normally lines the acini in a lobule. A contiguous epithelial cell layer and a basilar, intermittent myoepithelial cell layer. Hyperplasia, defined as an increase in the number of cells, is present when there are three or more cells above the basement membrane. In both ALH and LCIS, a monomorphic cell population fills, distends, and distorts the acini in the lobular unit. In ALH, filling of the acini is incomplete, other cell types may be intermixed with the monomorphic cell population, and fewer than half of the acini in the lobular unit are expanded and distorted by the proliferating cells. In LCIS, the acini are filled with a monomorphic cell population and at least half of the acini are distended and distorted by the proliferating cells. In some patients the distinction between LCIS and ductal carcinoma in situ (DCIS) may be difficult for the pathologist. Immunohistochemical staining for E-cadherin, a cell adhesion molecule, is used in some patients to distinguish lobular lesions that do not stain from ductal lesions that do stain for E-cadherin.

Lobular neoplasia is an uncommon diagnosis, reported in 0.5% to 3.8% of benign breast biopsies. It is diagnosed predominantly in premenopausal women and is characterized as a multicentric and bilateral process. Although there are now reported cases of calcifications identified in foci of lobular neoplasia, this is the exception. In most patients, lobular neoplasia is an incidental finding in biopsies done for palpable or mammographic findings. Women with lobular neoplasia are at increased risk for developing invasive ductal or lobular carcinoma within the first 10 to 15 years following the diagnosis. The increased risk reportedly applies to both breasts, though more recently there has been a report suggesting that the risk is higher in the breast diagnosed with the lobular neoplasia.

It is postulated that lobular neoplasia regresses following menopause and that these processes are marker lesions for increased risk of subsequently developing breast cancer. However, unlike DCIS, which is thought to progress to invasive disease in some women, the traditional teaching has been that lobular neoplasia is not precancerous. It is interesting to note, however, that in close to 50% of postmenopausal women diagnosed with invasive lobular carcinoma, prominent lobular neoplasia is diagnosed in association with the invasive lesion. This seems to challenge the notion that this is not a precancerous lesion and that it regresses in all patients. Alternatively, it may be that this lesion recurs in some patients.

The management of patients with lobular neoplasia diagnosed incidentally on core biopsy is evolving and remains controversial. Unfortunately, available studies at this time are limited by the relatively low number of patients reported and the potential bias built into the retrospective nature of the studies. In the past, excisional biopsy was not usually recommended for most of these patients. More recently, some investigators have suggested that excision is required if there is an overlap in the histologic features between LCIS and DCIS, if the histologic and imaging findings are discordant (e.g., if lobular neoplasia is all that is reported histologically, this may be an inadequate explanation for the findings prompting the biopsy), or in those patients in whom the lobular neoplasia coexists with another high-risk lesion. Alternatively, a growing number of authors suggest that excision should be the recommendation for this small group of patients because available reports in the literature describe malignancy in 0% to 50% of patients in whom excision is recommended following a core biopsy with incidentally noted lobular neoplasia.

 

Patient 21

Figure 4.21. Screening study, 76-year-old woman. Craniocaudal (A) and mediolateral oblique (B) views. Craniocaudal (C) and mediolateral oblique (D)views, right breast. Craniocaudal (E) and mediolateral oblique (F) spot compression views, right breast.

 

What do you think?

Is this a normal study, or is additional evaluation indicated?

A possible area of distortion is imaged in the upper outer quadrant of the right breast. Additional evaluation is indicated.

BI-RADS® category 0: need additional imaging evaluation.

 

What do you think, and what is your differential?

The spot compression views confirm the presence of distortion. The finding is readily apparent on the craniocaudal view and although it is identifiable on the oblique spot compression view, it is less striking. Differential considerations include fat necrosis related to prior surgery or trauma, complex sclerosing lesion, sclerosing adenosis, papilloma, focal fibrosis, an inflammatory process, invasive ductal carcinoma not otherwise specified, tubular carcinoma, and invasive lobular carcinoma. Rarely, ductal carcinoma in situ can present with distortion in the absence of calcifications. On physical examination, no scar is identified at the expected location of the distortion and no tenderness is elicited with compression. Given the Leborgne sign, which is that on palpation invasive ductal carcinomas are larger than what is seen on the mammogram, one would expect this lesion to be palpable, but it is not. On ultrasound, no abnormality is seen at the expected location of the mammographic finding. This suggests either a complex sclerosing lesion or an invasive lobular carcinoma. Atypical ductal hyperplasia is reported on core biopsies. Excisional biopsy is recommended, and a complex sclerosing lesion with atypical ductal hyperplasia and sclerosing adenosis is reported on the excised tissue.

The management of patients with complex sclerosing lesions remains controversial. With respect to these lesions, we need to consider several related questions. On core samples, can the pathologist reliably distinguish a complex sclerosing lesion from tubular carcinomas and sclerosing adenosis? In a patient with distortion and spiculation, can we accept the diagnosis of a complex sclerosing lesion after core biopsy? Is the association of complex sclerosing lesions with other lesions, including atypical ductal hyperplasia (ADH), lobular neoplasia, low-nuclear-grade ductal carcinoma in situ, and tubular carcinomas, frequent enough to warrant excisional biopsy in all patients with complex sclerosing lesions? If a complex sclerosing lesion is suspected based on imaging and clinical findings, should an imaging-guided core biopsy be done?

In most patients, a complex sclerosing lesion can be diagnosed on core biopsy samples. Rarely, the distinction among this entity, sclerosing adenosis, and tubular carcinomas can be a challenge histologically. Because atypical ductal hyperplasia, lobular neoplasia, ductal carcinoma in situ (usually low- or intermediate-grade), or tubular carcinomas are reported in as many as 33% of patients with complex sclerosing lesions, I recommend excision of all complex sclerosing lesions. Alternatively, some have suggested that if the lesion is not associated with ADH, the biopsy included at least 12 specimens, and the mammographic findings are reconciled with the histologic findings, no excision is required.

Complex sclerosing lesions can demonstrate fairly distinctive mammographic, sonographic, and clinical findings. The mammographic findings for these lesions include distortion with central 1-to 2-mm locules of fat (i.e., no significant central density), long curvilinear spicules, and better visualization in one of the two projections, commonly the craniocaudal view. Round and punctate calcifications may be seen in as many as 30% to 40% of lesions. These lesions are not related to a prior biopsy and, although they are occasionally palpable, most have no associated clinical finding (unlike what would be expected for an invasive ductal carcinoma of comparable size). On ultrasound, normal tissue or a subtle area of distortion with some shadowing that is not necessarily confirmed on the orthogonal image may be noted.

Based on imaging and clinical features, the likelihood of a complex sclerosing lesion can be predicted in a high percentage of patients. In patients in whom I consider the likelihood of a complex sclerosing lesion to be high, I recommend an excisional biopsy and forgo the imaging-guided biopsy. For those patients in whom a complex sclerosing lesion is in the differential but the imaging and clinical findings are not diagnostic, I do an imaging-guided biopsy; if a complex sclerosing is reported on the cores, I recommend excisional biopsy.

Patient 22

Figure 4.22. First screening study, 39-year-old woman. Craniocaudal (A) and mediolateral oblique (B) views. Mediolateral oblique (C), anterior compression views obtained as part of her screening study. Multiple spot compression (D–G) views of the findings in the left breast. Ultrasound image (H), radial projection of the mass in the upper outer quadrant of the left breast. Specimen radiograph (I), confirming excision of mass and adjacent area of distortion. Specimen radiograph (J), photographically coned to area of distortion.

Is this a normal or potentially abnormal mammogram?

What are the pertinent observations?

An oval mass is imaged in the upper outer quadrant of the left breast. Are there any other observations? How about a possible area of distortion at the posterior edge of the obvious mass in the left breast?

BI-RADS® category 0: need additional imaging evaluation.

 

 

How would you describe the findings, and what is your differential?

An area of distortion is confirmed on the spot compression views (Fig. 4.22D–F). It is characterized by the presence of fatty tissue centrally and long radiating spicules. It is more apparent on the craniocaudal view (Fig. 4.22D). Differential considerations include postsurgical or traumatic change, complex sclerosing lesion, papilloma, inflammatory change, focal fibrosis, sclerosing adenosis, fibromatosis, invasive ductal carcinoma not otherwise specified, and tubular carcinoma. Rarely, ductal carcinoma in situ can present with distortion in the absence of calcifications. Invasive lobular carcinoma is also in the differential, however, given the patient's age this is less likely. The patient has no history of breast surgery or trauma, and she is otherwise asymptomatic. Her physical examination in the expected location of the area of distortion is normal. No abnormality is identified on ultrasound corresponding to the area of distortion seen mammographically. A complex sclerosing lesion is suspected and excisional biopsy is recommended.

BI-RADS® category 4: suspicious abnormality, biopsy should be considered.

An oval mass with partially obscured and indistinct margins is confirmed on the spot compression views (Fig. 4.22G). Differential considerations for this finding include cyst, fibroadenoma (complex fibroadenoma, tubular adenoma), phyllodes tumor, papilloma, pseudoangiomatous stromal hyperplasia, adenosis tumor, and focal fibrosis. Depending on the clinical context, galactocele, postoperative or traumatic fluid collection, and an abscess are also in the differential. Invasive ductal carcinoma not otherwise specified and medullary carcinoma are the main considerations in the malignant category. Mucinous and papillary carcinoma are less likely, given the patient's age.

On ultrasound, a well-circumscribed mass with a heterogeneous echotexture and associated cystic areas is imaged corresponding to the mass seen mammographically (Fig. 4.22H). Biopsy is indicated.

BI-RADS® category 4: suspicious abnormality, biopsy should be considered.

The specimen radiograph is used to confirm excision of the mass and adjacent area of distortion (Fig. 4.22I, J). The localizing wire is seen in the specimen between the two lesions. Based on this image, the location of the mass and distortion are marked for the pathologist. A complex sclerosing lesion with no associated proliferative lesions is diagnosed for the area of distortion, and an adenosis tumor with significant fibrosis is reported for the mass.

Complex sclerosing lesions can demonstrate fairly distinctive mammographic, sonographic, and clinical findings. The mammographic findings for these lesions include distortion with central 1- to 2-mm locules of fat (i.e., no significant central density), long curvilinear spicules, and better visualization in one of the two projections, commonly the craniocaudal view. Round and punctate calcifications may be seen in as many as 30% to 40% of lesions. These lesions are idiopathic and are not related to a prior biopsy or trauma. Unlike the palpable findings one would expect to find for a comparably sized invasive ductal carcinoma, complex sclerosing lesions are usually not palpable. On ultrasound, normal tissue or a subtle area of distortion with some shadowing that is not necessarily confirmed on the orthogonal image may be noted.

For patients in whom, based on imaging and clinical findings, I suspect a complex sclerosing lesion, I recommend excision so that the lesion can be evaluated in its entirety. In nearly 33% of patients with complex sclerosing lesions, atypical ductal hyperplasia, lobular neoplasia, ductal carcinoma in situ (usually low- or intermediate-grade), or tubular carcinoma is reported being associated with the lesion. For patients in whom a complex sclerosing lesion is in the differential, but the imaging and clinical findings are not diagnostic, I do an imaging-guided biopsy; if a complex sclerosing is reported, I recommend excisional biopsy.

Alternatively, some recommend doing imaging-guided core biopsies on all of these lesions. If the lesion is a breast cancer, the patient can then have definitive surgery. If a complex sclerosing lesion is diagnosed on the cores, some recommend excisional biopsy. Others have suggested that if the complex sclerosing lesion is not associated with ADH, the biopsy includes at least 12 specimens, and the mammographic findings are reconciled with the histologic findings, no excision is required.

 

Patient 23

Figure 4.23. Screening study, 77-year-old woman. Craniocaudal (A) and mediolateral (B) oblique views. Craniocaudal (C) and mediolateral oblique (D)spot compression magnification views, right breast. When positioning for the additional views, the technologist notes nipple discharge. Ultrasound images in the subareolar area (E–G) and at the 2 o'clock position, 4 cm from the right nipple (H). Ultrasound image (I) demonstrating the round solid mass with associated spicular echoes consistent with the calcifications seen mammographically (thin arrow) and a second adjacent, irregular solid mass (thick arrow) at the 2 o'clock position, 4 cm from the right nipple. Specimen radiograph (J) demonstrating multiple masses, one of which has associated calcifications in the specimen. The preoperative wire localization is done using ultrasound guidance. Two wires are used to bracket the location of the intraductal lesions (seen on ultrasound) and the more peripheral cluster of masses seen mammographically.

 

Is this a normal study, or is additional evaluation indicated?

A mass with associated calcifications is present medially in the right breast. Additionally, at least on the craniocaudal view, nodularity is noted extending from the mass with calcifications towards the nipple.

BI-RADS® category 0: need additional imaging evaluation.

 

How would you describe the findings, and what would you do next?

A cluster of masses, one of which is associated with pleomorphic calcifications, is confirmed on the spot compression magnification views obtained of the right breast. A biopsy is indicated. An ultrasound is done to determine if the lesion can be identified for imaging-guided biopsy. Because of the nipple discharge noted by the technologist, additional history is obtained from the patient and, if indicated, physical examination and a ductogram are done.

On questioning, the patient relates having noticed her nipple dripping and dark brown spots on her bra and night clothes. On physical examination, nipple discharge is elicited easily from a single duct opening. A ductogram (not shown) and ultrasound are done for further evaluation.

Multiple intraductal masses are imaged in the subareolar area in moderately dilated ducts. These are confirmed on the ductogram (not shown). In addition, at least two solid masses, one of which has associated calcifications (Fig. 4.23I, thin arrow), are imaged at the 2 o'clock position, 4 cm from the right nipple. Given the clinical and imaging findings, multiple papillomas and associated invasive ductal carcinoma with ductal carcinoma in situ are the main considerations.

BI-RADS® category 4: suspicious abnormality, biopsy should be considered.

An ultrasound-guided biopsy of the mass with associated calcifications is done. An invasive ductal carcinoma with associated ductal carcinoma in situ, high nuclear grade with central necrosis, is reported on the core biopsy.

On the day of the lumpectomy, two wires are used to bracket the intraductal lesions seen on ultrasound, close to the nipple and the more peripheral cluster of masses. A 0.9-cm, grade III invasive ductal carcinoma with associated high-nuclear-grade ductal carcinoma with central necrosis is confirmed in the lumpectomy specimen. A 0.9-cm, high-nuclear-grade ductal carcinoma in situ with apocrine features is noted arising in an adjacent papilloma. Three excised sentinel lymph nodes are normal [pT1b, pN0, pMX, Stage I].

The management of patients diagnosed with papillomas on core needle biopsy remains controversial. Clearly, papillary lesions with atypia on core biopsy require excisional biopsy. The controversy centers on the diagnosis of a benign papillary lesion with no associated atypia on core biopsy. Many authors advocate that these can be followed with no excision required, while others recommend excision of all papillary lesions regardless of associated atypia. Among patients who have papillomas with associated atypia on core biopsy, 31% to 60% have been reported to have malignancy on excision. Among patients initially diagnosed with a papilloma (with no atypia), 0% to 18% have been reported to have malignancy on excision.

In considering the literature on papillary lesions, I think it is important to recognize the relatively low number of patients described, the retrospective nature of some of the studies, and the lack of adequate classification of lesions into central, usually solitary papillomas, and multiple peripheral papillomas. Additionally, the follow-up on many of these patients is limited; 2 or 3 years of follow-up is probably insufficient to truly assess the biologic significance of many of these lesions.

Patients with centrally occurring papillomas commonly present describing nipple discharge. The papillomas are often solitary, and on excision, no significant proliferative changes are typically reported in the surrounding breast parenchyma. In contrast, multiple peripheral papillomas are usually detected mammographically as one or multiple masses that may have associated calcifications or as multiple clusters of pleomorphic calcifications. Associated proliferative changes that include atypical ductal hyperplasia, lobular neoplasia, ductal carcinoma in situ (usually low-nuclear-grade), and invasive ductal carcinoma can be seen in nearly 50% of patients with multiple peripheral papillomas. Also, following excisional biopsy of these lesions, many patients present with recurrent lesions (new masses, or calcifications at the prior site). In the context of multiple peripheral papillomas, the term papillomatosis needs to be considered. This is a confusing term because some pathologists use it to describe multiple peripheral papillomas (i.e., lesions with a central fibrovascular core) and others use it for intraductal hyperplasia. I think it is best to avoid the term; however, when I am confronted with it, I specifically ask the pathologist how he or she is using the term.

As with so many other situations in breast imaging, the clinical context is important in determining the appropriate management of patients with papillary lesions. The larger the lesion, the greater the number of findings, and the older the patient, the more appropriate an excisional biopsy seems when a papilloma is diagnosed on core biopsy. It may be that patients with multiple peripheral papillomas should be treated more aggressively and that excisional biopsy may be appropriate in this patient population following an imaging-guided biopsy. In this context, it is also important to recognize that distinguishing benign papilloma from papillomas with atypia and papillary carcinoma may be difficult for the pathologist given core samples. As with the distinction between normal breast ductules and tubular carcinoma, it is the presence or absence of myoepithelial cells that distinguishes benign from malignant papillary lesions, and as much as 10% of a malignant papillary lesion has myoepithelial cells present, introducing the possibility of sampling bias.

 

 

Patient 24

Figure 4.24. Diagnostic evaluation, 71-year-old woman called back for evaluation of calcifications in the left breast detected on her screening study. Craniocaudal (A) and mediolateral oblique (B) spot compression magnification views. Specimen radiography (C) obtained following wire localization of the breast lesion.

What would you recommend?

A cluster of pleomorphic calcifications with round, punctate, and amorphous forms is confirmed on the magnification views. No definite linear forms are present, and there is no linear orientation. Fibrocystic changes including hyperplasia, atypical ductal hyperplasia, columnar alteration with prominent apical snouts and secretions (CAPSS), and sclerosing adenosis, as well as fibroadenoma, papilloma, and ductal carcinoma in situ (usually low- or intermediate-nuclear-grade, with no central necrosis) are in the differential for this cluster of calcifications. A stereotactically guided needle biopsy is recommended; however, the patient is short of breath and unable to lie prone. Excisional biopsy is scheduled following wire localization of the cluster of calcifications.

The specimen radiograph is obtained for several reasons. Confirmation that the localized lesion is excised is the primary reason, and although the specimen is a two-dimensional representation of a three-dimensional structure, if the localized abnormality is in close proximity to the margins, it is equally important that this is communicated to the surgeon. In evaluating the specimen, you also want to make sure that the localization wire has been removed with the specimen. Rarely, additional unsuspected lesions may be detected on the specimen; and lastly, the location of the excised lesion in the specimen is marked for the pathologist to assure that the excised lesion is evaluated histologically.

For this patient, atypical ductal hyperplasia (ADH) is reported histologically on the excised tissue. No further intervention is recommended; however, a mammogram of this breast in 6 months is requested, to establish a new baseline for this patient. Our approach to patients diagnosed with a high-risk marker lesion (e.g., ADH, lobular carcinoma in situ, atypical lobular hyperplasia, papilloma with atypia, multiple peripheral papillomas, complex sclerosing lesions, CAPSS with atypia, and mucocele-like lesions) is to obtain a mammogram following the excisional biopsy so as to document post biopsy changes as they peak during the first 6 months following the biopsy. After the first 6 months, postbiopsy changes stabilize or slowly resolve. By evaluating the patient 6 months following the surgery, it is unlikely that a spiculated mass at the biopsy site represents interval development of an invasive lesion. Most important, we avoid finding ourselves with a spiculated mass at the biopsy site (of a high-risk lesion) 2 years or more after the surgery, and not knowing if this is postoperative change that is regressing or the development of a more significant lesion.

 

 

Patient 25

Figure 4.25. Diagnostic evaluation, 61-year-old woman called back for evaluation of calcifications detected in the right breast on her screening mammogram. Craniocaudal (A) and mediolateral oblique (B) double spot compression magnification views, right breast. Specimen radiograph (C), obtained to confirm excision of the localized lesion and the wire.

 

How would you define these findings?

An area of amorphous calcifications is demonstrated on the magnification views. Differential considerations include fibrocystic changes, hyperplasia, atypical ductal hyperplasia, sclerosing adenosis, columnar alteration with prominent apical snouts and secretions (CAPSS), fibroadenoma, papilloma, and ductal carcinoma in situ (usually low-nuclear-grade with no central necrosis).

BI-RADS® category 4: suspicious abnormality, biopsy should be considered.

Following preoperative wire localizations, the specimen is placed in a container that allows for compression of the specimen with an alphanumeric grid. Although we apply compression on the specimen, we try to minimize the amount of compression because recent reports in the literature suggest that vigorous specimen compression may result in “false positive” histologic interpretations of tumor extending to the margins. The specimen radiograph is done using magnification technique on a mammographic unit, or a dedicated specimen radiography unit. The use of an alphanumeric enables us to place a pin through the center of the lesion, or four pins can be used to delineate the margins of larger lesions for the pathologist. This assures that the area of mammographic concern is fully evaluated histologically. Extensive sclerosing adenosis with associated calcifications is reported histologically.

Are the imaging and histologic findings congruent?

What do you recommend for this patient?

Yes, the imaging and histologic findings are congruent. The excised calcifications are identified as being associated with sclerosing adenosis, and no atypia or other high-risk lesion (e.g., complex sclerosing lesion or papilloma) is described by the pathologist. The patient is asked to return in 1 year.

Calcifications with this appearance typically occur in women with dense glandular tissue and may demonstrate a focal but loosely clustered distribution, as in this patient, or the calcifications may be diffusely scattered bilaterally. The differential to consider includes sclerosing adenosis, columnar alteration with prominent apical snouts and secretions (CAPSS), hyperplasia, atypical ductal hyperplasia, fibroadenoma, papilloma, and ductal carcinoma in situ (commonly low- or intermediate-grade with no associated central necrosis). The management of patients with this calcification type, particularly when diffusely involving the breast parenchyma bilaterally, can be a challenge. In women with focal findings, particularly if the calcifications represent a change compared with prior studies or they are in an unusual location (e.g., inner quadrants as opposed to the upper outer quadrant of the involved breast), I recommend a stereotactically guided biopsy. Patients with more diffuse and bilateral findings pose more of a management issue.

Do you sample or not?

If you sample, how do you decide where?

If you sample and the pathology is benign, can you be sure it is representative of all the calcifications?

My approach to patients with bilateral, diffusely scattered amorphous calcifications is annual follow-up with magnification views; I recommend biopsy when changes in the calcifications or the surrounding tissue are perceived on follow-up studies.

Sclerosing adenosis is a component of fibrocystic change most commonly seen in the perimenopausal period. It is a lobulocentric lesion characterized by disordered proliferation of epithelial, myoepithelial, and stromal elements. This process is characterized by an increased number of acini that are compressed and obliterated by the proliferating intralobular stroma predominantly in the center of lobules. More cystically dilated acini are seen at the periphery of the involved lobules. Sclerosing adenosis is also characterized by hyperplasia of the myoepithelial cells. It can occur as an isolated lesion or as a component of complex sclerosing lesions, papillomas, fibroadenomas (e.g., complex fibroadenomas), and invasive or in situ cancers.

The imaging features of sclerosing adenosis are variable. Sclerosing adenosis may present as a palpable or screen-detected mass, the margins of which may range from well circumscribed to indistinct to spiculated; it may also present as distortion or focal parenchymal asymmetry, and associated calcifications may be present. Alternatively, sclerosing adenosis may present with calcifications and no associated mass. One of three patterns may be seen relative to calcifications developing in areas of sclerosing adenosis: a tight cluster of round and punctate, well-defined calcifications; loosely clustered amorphous calcifications; or bilateral, diffusely distributed amorphous calcifications in the setting of dense glandular tissue.

 

Patient 26

Figure 4.26. Screening study, 53-year-old woman. Right craniocaudal (A) and mediolateral oblique (B) views. Spot compression magnification craniocaudal view (C), right breast. Right craniocaudal (D, F, H, J) and mediolateral oblique (E, G, I, K) views, at 6, 18, 30, and 42 months following completion of the initial treatment. Linear metallic marker seen on some of the images is used to indicate the site of the lumpectomy scar. Right craniocaudal (D, F, H, J) and mediolateral oblique (E, G, I, K) views, at 6, 18, 30, and 42 months following completion of the initial treatment. Linear metallic marker seen on some of the images is used to indicate the site of the lumpectomy scar. Right craniocaudal (D, F, H, J) and mediolateral oblique (E, G, I, K) views, at 6, 18, 30, and 42 months following completion of the initial treatment. Linear metallic marker seen on some of the images is used to indicate the site of the lumpectomy scar. Right craniocaudal (D, F, H, J) and mediolateral oblique (E, G, I, K) views, at 6, 18, 30, and 42 months following completion of the initial treatment. Linear metallic marker seen on some of the images is used to indicate the site of the lumpectomy scar. Right craniocaudal (L) and mediolateral oblique (M) views, 11 months following the films shown in (J) and (K). The patient now describes a “lump” at the lumpectomy site. Spot compression magnification craniocaudal view (N), right breast. The patient now describes a “lump” at the lumpectomy site. Spot compression magnification craniocaudal view (N), right breast. Ultrasound images, transverse (TRV) (O) and longitudinal (LON) (P) projections of the mass in the right breast at the 1 o'clock position, 8 cm from the nipple.

How would you describe these findings, and what would you recommend next?

A cluster of calcifications is detected medially in the right breast. Spot compression magnification views are indicated for further evaluation.

BI-RADS® category 0: need additional imaging evaluation.

The magnification views (the mediolateral oblique projection is not shown) confirm the presence of a cluster of pleomorphic calcifications in the right breast. Ductal carcinoma in situ (DCIS) is the main differential consideration for these calcifications, and therefore a biopsy is indicated. Because of the density associated with the calcifications, invasion may be present. DCIS is diagnosed following core biopsies, and is confirmed on the lumpectomy. In addition to the lumpectomy, the patient is treated with radiation therapy to the right breast.

Postlumpectomy changes and surgical clips are present in the right breast at the lumpectomy site. Immediately following the completion of the radiation therapy (Fig. 4.26D, E), there is increased density at the lumpectomy site. The density progressively resolves, and oil cyst formation is noted at the lumpectomy bed. What do you think about the findings in the last set of films (Fig. 4.26J, K)? How would you report this study, and what is your recommendation?

At this time, there is an irregular 2-cm mass with indistinct margins, shadowing, and associated pleomorphic and linear calcifications, some of which demonstrate a linear orientation. The constellation of findings is consistent with recurrent disease. Ultrasound-guided biopsy is done and confirms the presence of invasive ductal carcinoma and associated high-nuclear-grade ductal carcinoma in situ (DCIS) with central necrosis. In retrospect, a density is seen developing at the lumpectomy bed on the last set of images (Fig. 4.26J, K), obtained 42 months following completion of treatment. A simple mastectomy with axillary dissection is done. A 3-cm invasive ductal carcinoma with associated high-nuclear-grade DCIS and lymphovascular space involvement is reported histologically. No metastatic disease is identified in two excised sentinel lymph nodes [pT2, pN0(sn) (i>2), pMX; Stage II].

Local recurrence (i.e., ipsilateral breast tumor recurrence, IBTR) following breast-conserving therapy or mastectomy is defined as the development of cancer in remaining ipsilateral breast tissue or skin or on the ipsilateral chest wall or skin, respectively. Regional recurrence is defined as the development of cancer in remaining ipsilateral axillary lymph nodes, supraclavicular, infraclavicular, or internal mammary lymph nodes.

Follow-up protocols for patients after lumpectomy and radiation therapy are variable. Some facilities follow patients with a history of conservatively treated breast cancer at 6-month intervals for 3, 5, or 7 years, and the contralateral breast at yearly intervals. Other facilities obtain annual mammograms bilaterally on these patients. We recommend annual diagnostic mammography for these patients and obtain routine craniocaudal and mediolateral oblique views bilaterally as well as a spot magnification tangential view of the lumpectomy site for the first 7 years following the surgery, after which we return the patient to screening. The development of new pleomorphic calcifications, a mass, or increasing density and distortion at or close to the lumpectomy site are mammographic findings that may be associated with a recurrence. Less commonly, recurrences may be characterized by diffuse breast changes, including a change in the size of the breast, increased density of the parenchyma, prominence of the trabecular markings, and skin thickening.

Reported recurrence rates range from 5% to 19% in the first 5 to 12 years following lumpectomy with radiation therapy, and from 4% to 14% following mastectomy. Risk factors linked to recurrence following lumpectomy with radiation therapy include young age at the time of presentation, extensive intraductal component, multifocal disease, lymphovascular space involvement, large tumor size, high histologic grade, tumor necrosis, and positive margins at the time of the original resection. Most recurrences occur at or near the site of the original tumor within the first 5 years following treatment.

Postlumpectomy changes are variable but may include areas of increased density, distortion, and spiculation at the site of the lumpectomy, often associated with skin thickening and distortion. These changes are usually most prominent within the first year following the surgery and then stabilize or progressively resolve. In some patients, oil cyst formation and the development of dystrophic calcifications may be seen at the lumpectomy site as the area of density and distortion decreases. Some patients develop postoperative fluid collections; these may also stabilize or progressively decrease in size, sometimes resolving completely without the need for any intervention. Radiation therapy changes more commonly involve the entire breast and include increased density and prominence of the trabecular markings as well as diffuse skin thickening. Radiation therapy changes typically resolve within the first 2 years following completion of the therapy.

 

 

Patient 27

Figure 4.27. Diagnostic evaluation, 56-year-old patient presenting with a “lump” in the left breast. Craniocaudal (A) and mediolateral oblique (B) views. Exaggerated craniocaudal (C) spot compression view of palpable finding, left breast. Ultrasound images, radial (RAD) (D) and antiradial (ARAD) (E)projections of palpable mass at the 2 o'clock position, posteriorly (Z3) in the left breast. Ultrasound images, radial (RAD) (F) and antiradial (ARAD) (G)projections, left axilla. Ultrasound images, radial (RAD) (H) and antiradial (ARAD) (I) projections, of the mass at the 2 o'clock position, posteriorly (Z3) in the left breast. Ultrasound image (J) documenting needle positioning in the center of the mass prior to deployment of the clip. Craniocaudal (K) and 90-degree lateral (L) views of the left breast, photographically coned, documenting clip location. Diagnostic evaluation, patient presents describing a “lump” at the lumpectomy site, 6 months following neoadjuvant therapy and, more recently, lumpectomy and radiation therapy. Craniocaudal (M) and mediolateral oblique (N) views, left breast. Spot compression (O) view of palpable finding. Ultrasound image (P) of mass at the lumpectomy site, 2 o'clock position, left breast.

How would you describe the findings, and what is your recommendation?

On physical examination, a hard fixed mass is palpated at the site of concern to the patient. A solid, 2-cm mass with indistinct margins and spiculation, associated linear calcifications, and areas of shadowing and enhancement is imaged corresponding to the palpable finding. Given a round mass, with some enhancement and associated linear calcifications, the histologic diagnosis can be predicted fairly accurately. The clinical and imaging findings are consistent with a poorly differentiated invasive ductal carcinoma with associated ductal carcinoma in situ with central necrosis (probably high nuclear grade). In patients with a high likelihood of a malignancy, the remainder of the breast and the ipsilateral axilla are scanned to try and identify multifocal or centric disease and possibly abnormal axillary lymph nodes.

 

 

How would you describe the findings, and what is your recommendation?

A potentially abnormal lymph node is identified in the ipsilateral axilla. An adjacent normal-appearing lymph node is also noted. Potentially abnormal lymph nodes are characterized by absence or marked attenuation of the echogenic hilar region in conjunction with thickening and bulging of the hypoechoic cortex. In some patients, potentially abnormal lymph nodes are nearly anechoic and no hyperechoic hilar region is present. When a potentially abnormal lymph node is identified, we recommend fine-needle aspiration or, depending on the size and location of the lymph node, a core biopsy (if it can be done safely). In this patient, core biopsies are done on the mass in the breast and of the potentially abnormal axillary lymph node. An invasive ductal carcinoma with associated ductal carcinoma in situ and metastatic disease to the axillary lymph node are reported on the core biopsies. With the diagnosis of metastatic disease to an axillary lymph node, the patient will undergo a full axillary dissection (i.e., sentinel lymph node biopsy is not indicated).

Treatment options discussed with this patient include mastectomy and axillary dissection followed by chemotherapy; lumpectomy with axillary dissection followed by radiation therapy and chemotherapy; or, being used with increasing frequency, neoadjuvant therapy followed by either mastectomy or lumpectomy, an axillary dissection, and radiation therapy for those patients having a lumpectomy. If a mastectomy is done following neoadjuvant therapy, radiation therapy is not always recommended. Neoadjuvant therapy can decrease the size of, or eliminate, the tumor (i.e., downstage the lesion), enabling some patients, who might not otherwise be candidates for conservative therapy, to undergo lumpectomy. This patient elects to undergo neoadjuvant therapy.

After two courses of chemotherapy, the mass has decreased in size (Fig. 4.17H, I). In patients who undergo neoadjuvant therapy, the initial findings may resolve completely. Consequently, in patients who may elect to have a lumpectomy after neoadjuvant therapy, marking the location of the original lesion is important so that this area can be localized, excised, and evaluated for residual disease. Given the significant response in this patient after two courses of chemotherapy, and with a planned lumpectomy if the tumor continues to respond, a “marking” clip is placed in the mass using ultrasound guidance (Fig. 4.27J).

 

Clips to mark the location of a lesion are deployed at the time of imaging-guided biopsies when a lesion (e.g. a cluster of calcifications or a small mass being biopsied with an 11G vacuum-assisted device), which may be malignant, might be removed in its entirety as a result of the biopsy. In these patients, if a malignancy is diagnosed and the original lesion has been removed completely, the clip deployed at the time of imaging-guided biopsy marks the site of the lesion. At the time of the lumpectomy, the clip is localized so that the tissue surrounding the original lesion can be evaluated histologically for residual disease. Similarly, if it is known that a patient will undergo neoadjuvant therapy, and because the lesion may resolve as a result of the therapy, a clip can be deployed in the lesion at the time of the biopsy. If the lesion responds to therapy, the clip is localized preoperatively in those patients undergoing lumpectomy so that the tissue at the site of the treated lesion can be evaluated histologically for residual disease. Alternatively, as in this patient (because histology is not always predictable at the time of the biopsy and treatment choice is not always known), the clip can be deployed during the therapy in those patients in whom the original lesion is responding and may resolve completely prior to the surgery.

Following completion of her neoadjuvant therapy, the patient undergoes a lumpectomy and full axillary dissection. A 2-cm area of fibrosis, consistent with tumor regression, is described at the site of the clip. No residual tumor is identified. No metastatic disease is identified in 11 excised axillary lymph nodes. Several of the lymph nodes demonstrate areas of fibrosis consistent with tumor regression [ypTX, ypN0, ypMX]. As in this patient, approximately 15% to 20% of patients who undergo neoadjuvant therapy demonstrate complete response with no residual tumor identified histologically at the time of their surgical procedure. Reportedly, the subgroup of patients with a complete pathologic response has higher relapse-free survival and overall survival rates compared with patients with residual disease at the time.

 

 

How would you describe the findings, and what is your main consideration at this point?

The overall density of the breast is increased, which is consistent with radiation therapy effect. A round mass with partially well circumscribed and obscured margins is imaged at the lumpectomy site and corresponds to the site of concern to the patient. Although this could represent a recurrence, given a complete pathologic response to neoadjuvant therapy and the short time interval between her lumpectomy and the development of this mass, a postoperative fluid collection is the primary consideration. An ultrasound is done for further evaluation.

What is your diagnosis and recommendation?

On physical examination, a hard mass is palpated at the lumpectomy site. On ultrasound, a well-circumscribed, complex cystic mass with posterior acoustic enhancement is imaged corresponding to the palpable finding. During the real-time portion of the study, some of the spicular echoes and hyperechoic bands in the mass shift in position and appear to be “floating” in the mass, respectively. This is a postoperative fluid collection requiring no intervention unless a superimposed infection is suspected or it is causing significant discomfort. A high recurrence rate is associated with aspiration. It is critical, however, to reassure the patient that this is not a recurrence and that fluid collections are common following lumpectomy. We can expect that this will decrease in size and stabilize or resolve completely with time. Postoperative fluid collections typically have a complex cystic appearance on ultrasound. As in this patient, many can be characterized as predominantly cystic, with spicular echoes, septations, and mural nodules, whereas others have small cystic spaces in what otherwise appears to be a solid matrix.

 

Patient 28

Figure 4.28. Diagnostic evaluation, 8-year-old girl with a mass in the left subareolar area. Ultrasound image (A) of left subareolar area. Ultrasound image (B) of the right subareolar area for comparison.

What is the diagnosis and recommendation?

On physical examination, the left breast is more prominent than the right, and a readily mobile mass is palpated in the left subareolar area. On ultrasound, an irregular area of hypoechogenicity with indistinct margins is imaged in the left subareolar area. A smaller but similar-appearing area is imaged in the right subareolar area. This is consistent with premature, asymmetric breast bud development (a very similar appearance is seen when an ultrasound study is done in men with gynecomastia). No further intervention is warranted.

Breast masses in children and adolescents are often benign and include inflammatory conditions, premature breast bud development, cysts, and fibroadenomas. Malignant causes are rare but include metastatic disease (rhabdomyosarcoma, neuroblastoma, lymphoma) and secretory carcinoma (a primary breast malignancy characterized by large amounts of intra- and extracellular secretion and neoplastic cells with granular eosinophilic cytoplasm). Although initially called juvenile carcinoma because it was associated with childhood and adolescence, it can occur in patients of all ages. Care is required in the management of these patients, because surgical removal of a developing breast bud results in significant deformity or failure of normal breast development. Consequently, fine-needle aspiration is the procedure of choice if a neoplastic process is a serious consideration in this patient population.

 

Patient 29

Figure 4.29. Diagnostic evaluation, 45-year-old patient presenting with a painful “lump” in the right breast. Craniocaudal (A) and mediolateral oblique (B) views, right breast, with a metallic BB at the site of concern to the patient. Spot tangential (C) view of the palpable finding. Ultrasound images, radial (RAD) (D) and antiradial (ARAD) (E) projections of palpable finding, right breast. Ultrasound images of aspiration. Preaspiration image, documenting needle positioning in the cyst (F), postaspiration image demonstrating no residual abnormality (G) surrounding the needle, and image obtained following the injection of air (H). Ultrasound image following air injection (I). Air is seen as an echogenic line (arrows). Pneumocystogram films. Spot compression magnification views of airfilled cyst, craniocaudal (J) and mediolateral oblique (K) projections.

 

How would you describe the findings, and what would you recommend next?

On the routine views, a mass with obscured margins is imaged corresponding to the palpable finding. On the spot tangential view, the margins of the mass are partially well circumscribed, indistinct and obscured. Differential considerations include cyst, fibroadenoma (complex fibroadenoma, tubular adenoma), phyllodes tumor, papilloma, pseudoangiomatous stromal hyperplasia, and focal fibrosis. Depending on history and clinical findings, an inflammatory process, posttraumatic/postsurgical fluid collection, and a galactocele might also be in the differential. Malignant considerations include invasive ductal carcinoma not otherwise specified, medullary carcinoma, or metastatic disease. Although they are less likely given the patient's age, mucinous and papillary carcinomas are also in the differential. Correlative physical examination and an ultrasound are indicated for further characterization.

 

What is the diagnosis, and what would you recommend?

On physical examination, there is a hard tender mass palpated at the 12 o'clock position, 2 cm from the right nipple. On ultrasound, this is an anechoic mass with posterior acoustic enhancement consistent with a cyst. There is some irregularity of a portion of the wall. Following discussion with the patient, an aspiration is undertaken, primarily for symptomatic relief. A pneumocystogram is also planned for further evaluation.

After establishing an approach that allows me to advance the needle parallel to the transducer, I clean the skin and use lidocaine to anesthetize the skin. Then, using ultrasound guidance, I inject lidocaine in the tissue leading up to, but taking care to not go into, the lesion. I use ultrasound guidance for administering the anesthesia and for doing the aspiration, even in patients in whom the mass is palpable. Commonly, the advancing needle displaces the mass, or indents the wall, but does not penetrate into the mass (I think this explains many of the patients who present for evaluation of a palpable mass following attempted aspirations that yielded no fluid and yet we find a cyst corresponding to the palpable finding). By visualizing the trajectory of the advancing needle, I can gauge the amount of compression I need to apply to effectively immobilize the mass and the amount of controlled pressure I need apply with the needle so that the cyst wall is punctured. With the needle in the cyst, I pull the stylet out of the 20G spinal needle, attach a 10-mL syringe, and aspirate. I watch on real time as I aspirate, to be sure there is no residual abnormality postaspiration. Also, in some patients, the needle may need to be redirected (i.e., the tip of the needle is against the cyst wall) during the aspiration, to be sure that all the fluid is aspirated. At this point, if I am doing a pneumocystogram, I stabilize the needle and replace the fluid-filled syringe with one holding air (half of the volume of the aspirated fluid), and I then inject the air into the cyst (Fig. 4.29H). The air is imaged as an echogenic line (Fig. 4.29I, arrows).

In this patient, 4 mL of serous fluid is aspirated and no residual abnormality is seen following the aspiration. For the pneumocystogram, half of the volume of fluid aspirated is replaced with air (Fig. 4.29H, I) and spot compression magnification views of the aspirated cyst are obtained (Fig. 4.29J, K). Possible wall irregularity and thickening or intracystic lesions can be further evaluated on the pneumocystogram. In this patient, the wall of the cyst is smooth and well defined. No intracystic lesion or wall abnormality is identified. Annual screening mammography is recommended.

BI-RADS® category 2: benign finding.

Following cyst aspiration, I routinely inject air into the cyst cavity. Some have suggested that air injection following aspiration can reduce the incidence of cyst recurrence. The air does not hurt the patient, and if it is helpful in minimizing the likelihood of a recurrence, it can be beneficial. For patients for whom I am concerned about the presence of a mural or intracystic abnormality, spot compression magnification views of the cyst are done following the injection of air (i.e., a pneumocystogram), to further evaluate the wall of the cyst.

 

 

Patient 30

Figure 4.30. Diagnostic evaluation, 55-year-old patient called back for evaluation of a cluster of calcifications in the left breast detected on her screening mammogram. Double spot compression magnification view (A), craniocaudal projection. Ultrasound images (B, C) done in the upper outer quadrant of the left breast at the expected location of the calcifications. Ultrasound images (D, E) demonstrating echogenic foci (large arrows) in the tissue consistent with the presence of calcifications. Shadowing (small arrows) can be seen associated with some of the echogenic foci. Radiograph of three cores (F)obtained using ultrasound guidance.

What is your working diagnosis, and what is your recommendation?

The magnification views confirm a cluster of calcifications with linear forms having irregular margins, clefts, and demonstrating linear orientation. The main consideration with calcifications having these features is ductal carcinoma in situ with central necrosis, likely high nuclear grade. Biopsy is indicated.

As described previously (see discussion with Figs. 4.18 and 4.19), these types of calcifications are closely associated with the malignant cells. Targeting the calcifications targets the malignant cells. If calcifications are removed in the cores, the diagnosis is established. Consequently, I evaluate patients having a tight cluster of this type of calcifications with ultrasound. Although in some of these patients a mass is identified sonographically in association with the calcifications, the primary reason for doing the ultrasound is not to detect or further characterize the calcifications, but rather to determine whether I can use ultrasound guidance for the biopsy. My preference is to do ultrasound-guided core biopsies because patients are more comfortable in a supine position with no breast compression, and no radiation is needed. If I can see the calcifications with ultrasound, I can target them.

 

 

What do you think?

Multiple spicular echoes (large arrows, Fig. 4.30D, E) some with associated shadowing (Fig. 4.30D, small arrows) are imaged at the expected location of the calcifications seen mammographically in the upper outer quadrant of the left breast. Three 14G core samples are done through this area, and core radiographs are obtained. If calcifications are confirmed on the cores, nothing further is done (remember that with this type of calcification, removing calcifications targets the malignant cells directly). If no calcifications are seen, additional cores are obtained.

 

What do you think?

Calcifications are present in all of the cores. A high-nuclear-grade ductal carcinoma in situ with central necrosis is diagnosed on the cores.

My approach to patient care is to do whatever I need to do to arrive at the correct diagnosis as simply, easily, and efficiently as I can. This is in the context of complete, high-quality imaging workups; I am not suggesting or advocating cutting corners or accepting suboptimal work—quite the opposite. Although I am sure that percutaneous treatment of small breast cancers is the future, at this time that is not what I am trying to do. When I do a breast biopsy, I am trying to arrive at a correct diagnosis in the easiest, least invasive, and most efficient way possible for my patient. If the correct diagnosis can be established with a fine-needle aspiration or one or two cores using a 14G needle, why do more? If there is no chance that I will remove a lesion in its entirety during the biopsy, why deploy a clip? Just because I can? I challenge you to think methodically about what and how you do things to patients. Step back and ask yourself: What do I absolutely need to do to take care of this patient optimally, and what is the easiest way to accomplish the goal? Keep it simple!

 

Patient 31

Figure 4.31. Screening mammogram, 73-year-old woman. Craniocaudal (A) and mediolateral oblique (B) views, left breast, photographically coned to an area of calcifications. Craniocaudal (C) and 90-degree lateral (D) views from a preoperative localization done for a mass in the left breast 4 years prior to (A) and (B). Diagrams illustrating preoperative localization using a free-hand anteroposterior approach for needle/wire placement. With the breast uncompressed, the needle is advanced in the breast toward the expected location of the lesion. Craniocaudal and 90-degree lateral (E) views are done to establish the relationship of the needle to the lesion. Based on these initial images, the needle is repositioned and images are repeated (F). Because the needle is still not in close proximity to the lesion on orthogonal views, the needle is repositioned and another set of images (G) is done. Depending on your persistence, the needle may need to be repositioned several times before it is close enough to the lesion to assure adequate localization and excision. After the position of the needle is deemed adequate, the wire is deployed. Depending on the size and location of the lesion being localized, the size of the breast, and the experience of the breast imager with this technique, accurate localizations using this approach may be more difficult. Diagrams illustrating concepts of preoperative wire localization using breast compression with an alphanumeric grid for a parallel-to-chest-wall approach. Ninety-degree and craniocaudal views of the breast (H) are reviewed, and the distances from the various skin surfaces to the lesion are measured. The distance from the superior aspect of the breast to the lesion in this example is “x” cm, the distance from the inferior aspect of the breast to the lesion in this example is “y” cm, the distance from the lateral aspect of the breast in this example is “z” cm, and the distance from the medial aspect of the breast is “s” cm. The shortest distance from the skin to the lesion dictates the approach used for needle placement. In this example, the lesion is closest to the medial aspect of the breast, so a mediolateral approach is used. A needle long enough to go 1 cm beyond the lesion is selected. Using a compression paddle that has a central fenestration surrounded by an alphanumeric grid, a 90-degree mediolateral view is done (I). The coordinates for the center of this lesion are “B” and “3”. Using the collimator light, a shadow of these coordinates is cast on the patient's breast, lidocaine is used at the intersection point of these coordinates, and the needle is advanced in the breast. A second 90-degree mediolateral view is done (J). Because the wire is placed through the needle, this view describes the eventual trajectory and relationship of the wire to the lesion in this plane. The hub of the needle (black square superimposed on the mass) is superimposed on the lesion. Next, using the spot compression paddle, a craniocaudal view (J) is done to document the relationship of the needle to the lesion in the orthogonal projection. The breast is kept in compression in this projection until after the wire is deployed. The needle should be through or within 5 mm of the mass and 1 cm beyond the lesion. If the tip of the needle is more than 1 cm beyond the lesion, the needle is pulled out as much as needed for the tip to be 1 cm beyond the lesion. If the needle is associated with the lesion on the orthogonal views (as in this example), the wire is deployed and another craniocaudal view (K) is done to document final wire positioning. After the wire is deployed, the breast is not compressed again in a projection that is perpendicular to the wire (the lateral projection in this example). Ultrasound image (L) obtained after the wire is deployed in a hypoechoic mass in the left breast. If the trajectory of the needle (and consequently the wire) is parallel to the orientation of the transducer, the needle and wire can be seen in their entirety. In this patient, you can see the hook of the wire and a portion of the reinforced wire segment in the mass. The distance from the skin to the mass is measured, and using an indelible marker, an “X” is placed on the skin directly over the mass/wire as an additional guide for the surgeon. A mammographic image (M) obtained with compression applied parallel to the course of the wire is obtained to document the position of the wire in the lesion. A radiograph of the specimen (N) is obtained to verify excision of the localized lesion and localization wire and to mark the location of the lesion for the pathologist. Alternatively, sonography of the specimen can be done to document excision of the lesion and is indicated when the lesion is seen on ultrasound only (i.e., it is not seen mammographically).

What do you think?

Are magnification views indicated?

These are dense, coarse calcifications, most likely dystrophic in etiology. Magnification views are not indicated. In looking through the patient's jacket, prior studies provide an explanation for the calcifications. Vascular calcifications are also present.

 

What approach was used for the localization, and what are the limitations and possible complications inherent in this approach?

What are the alternative options for preoperative wire localizations?

In this patient, the calcifications are dystrophic and localized to the prior site. They require no further evaluation, intervention, or short-interval follow-up. Annual screening mammography is recommended for the patient.

A free-hand anteroposterior, or frontal, approach was used to localize the mass in the left breast. This is an acceptable method for preoperative localizations; however, it is more of a challenge to localize lesions precisely using this approach. Also, because the needle is advanced toward the chest wall, care should be exercised to minimize the possibility of a pneumothorax, particularly in thin patients with small breasts.

The main challenge in using this method is that it requires you to establish the location of a lesion in the breast, based on images with the breast compressed and pulled out away from the body, and transpose this to a breast that is uncompressed and in its natural position when the needle is advanced in the breast. After you position the needle in the breast free-hand, craniocaudal (Fig. 4.31C) and 90-degree lateral (Fig. 4.31D) views are obtained. In this patient the needle is through the lesion in both images (Fig. 4.31C, D). However, this degree of accuracy is difficult to obtain and often requires serial approximations. Depending on the relationship of the needle to the lesion on the initial images, free-hand adjustments are made to the position of the needle. The images are repeated and, based on the new position of the needle, additional adjustments may be indicated. This is done as many times as necessary to position the needle as close to the lesion as possible (Fig. 4.31E–G). Depending on the size of the breast, the size and location of the lesion being localized, and the experience and persistence of the breast imager, having the needle consistently through or within 5 mm of the lesion is hard to achieve with this method. The issue then becomes what you are willing to accept as an adequate position (distance) for the wire relative to the lesion: Is it acceptable if the wire is 1 cm or 1.5 cm from the lesion? Ideally, you want the wire to be within 5 mm of the lesion, and you do not want the wire to be short of the lesion.

A more accurate method involves using breast compression with an alphanumeric grid and an approach for needle placement that is parallel to the chest wall. This is a simple and safe approach that enables precise localization of even the smallest of lesions. Because the needle is advanced in the breast parallel to the chest wall, the possibility of a pneumothorax is eliminated. The possible routes for needle entry using this approach include craniocaudal, caudocranial (i.e., from below), lateromedial, or mediolateral. The shortest distance from the skin to the lesion is determined on craniocaudal and 90-degree lateral views (Fig. 4.31H) of the breast, and this dictates the route taken for needle entry. The shortest distance to the lesion on the example provided is “s” cm using a mediolateral approach for the localization. A needle that is long enough to go 1 cm beyond the lesion is selected (i.e., 1 cm >1 “s” cm).

The breast is positioned for a 90-degree mediolateral view using the alphanumeric grid to compress the medial aspect of the breast (Fig. 4.31I). The patient's breast is kept in compression after this image is taken. The coordinates for the lesion are established on the image (“B” and “3”) and using the collimator light (or laser light) a shadow of the lesion coordinates is cast on the breast. After anesthetizing the skin entry site with lidocaine, the needle is advanced in the breast at the intersection point for the coordinates. If the patient has not moved from the time the initial image is done to the time you introduce the needle, and you selected a needle long enough to go beyond the lesion, you will have skewered the lesion. At this point, we do another 90-degree mediolateral view (Fig. 4.31J) to document needle and, after deployment, wire positioning in this projection and release compression. A craniocaudal view using the spot compression paddle (Fig. 4.31J) is done next, and the breast is kept in compression after this view is obtained. On the 90-degree mediolateral view, the hub of the needle should be superimposed on the lesion, and on the CC view, the needle should be through or within 5 mm of the lesion and extend 1 cm beyond the lesion. If the needle is correctly positioned on the orthogonal views, the wire is advanced through the needle and the needle is pulled out, making sure that you do not inadvertently pull the wire out with the needle. A repeat CC view is done to document final positioning. Compression is released and the portion of the wire external to the breast is secured on the skin.

After the wire is deployed, we do not compress the breast in a perpendicular direction to the wire because this can result in unwanted changes in the position of the wire (e.g., the wire can be pulled out and end up short of the lesion, or the wire can be advanced significantly beyond the lesion). Because the wire is placed through the needle, the position of the needle in the initial projection (Fig. 4.31J, 90-degree mediolateral view in this example) describes the eventual position of the wire in this projection. There is no need, therefore, to repeat this projection (requiring you to compress the breast perpendicular to the wire) after the wire is deployed.

Lastly, preoperative wire localizations can be done using ultrasound guidance. My general rule is that if I can see the lesion with ultrasound, I prefer to use ultrasound guidance for biopsies and preoperative wire localizations. In order to see the needle (and subsequently the wire) in its entirety, I establish an approach that allows me to advance the needle in the breast parallel to the transducer. I use a 25G, 1.5-in needle to inject lidocaine at the skin entry site and in the expected trajectory of the needle up to the lesion. I then advance the needle through the breast and into the lesion and verify that the needle is through the lesion longitudinally and in cross section (via orthogonal ultrasound images of the needle). Preferably with the tip of the needle 1 cm beyond the lesion, I deploy the wire, remove the needle, making sure that I do not inadvertently pull the wire out with the needle, and obtain orthogonal ultrasound images of the wire (Fig. 4.31L). I measure the distance from the skin directly down to the location of the lesion/wire for the surgeon and I place an “X” on the skin surface directly over the lesion. A single mammographic image is obtained of the wire (Fig. 4.31M). The view used is selected so that compression of the breast occurs parallel (and not perpendicular) to the wire. After the wire is deployed, we do not compress the breast perpendicular to the wire because this might result in changing the final wire positioning (see Fig. 4.32). A radiograph or ultrasound of the specimen is always obtained following preoperative wire localizations to document excision of the localized abnormality (Fig. 4.31N).

 

 

Patient 32

Figure 4.32. Preoperative wire localization, 47-year-old patient. Craniocaudal (A) and 90-degree lateral (B) views. Diagram illustrating the inadvertent repositioning of the wire that can result when breast compression is applied perpendicular to the direction of wire deployment (C). This can have an accordion effect so that the wire is advanced in, or pulled out, of the breast. After compression is released, and the orthogonal view is obtained (D), the wire may have been advanced significantly beyond the lesion (I.), pulled out of the lesion (II.), or it may remain appropriately positioned (III.). While it is acceptable (though not desirable) to have the wire beyond the lesion, it is not acceptable for the wire to be short of the lesion (II.).

How was this localization approached, and what do you think about the final position of the wires?

In this patient, two lesions in the right breast (a clip deployed after a stereotactically guided biopsy is evident in one of the lesions) are localized using a 90-degree lateromedial approach. The shortest distance from the skin to the lesion being localized, as measured on craniocaudal (CC) and 90-degree lateral views, dictates the approach that is taken when the parallel-to-the-chest-wall approach using an alphanumeric grid is the method selected for preoperative wire localizations. In this patient, the lesions are closest to the lateral aspect of the breast on the CC view. Consequently, the needles/wires are placed in the breast correctly using a 90-degree lateromedial approach.

The problem with this wire localization is that the wires ended up significantly beyond the lesions, as seen on the CC view. This can occur if, after the wire is deployed, the breast is compressed perpendicular to the direction in which the wire is deployed (Fig. 4.32C, D). In this patient the lateral view (Fig. 4.32B) was done after the wires were deployed, resulting in the inadvertent repositioning of the wires.

An image is always done after the needle is placed in the breast in the initial projection (see Fig. 4.31J). Because the wire is placed through this needle, the position of the needle describes the eventual position of the wire in the initial projection. There is no need, therefore, to repeat this projection (requiring you to compress the breast perpendicular to the wire) after the wire is deployed.

 

 

Patient 33

Figure 4.33. Preoperative wire localization, 56-year-old patient. Ninety-degree lateral view (A) obtained after wire placement. Specimen radiograph (B).

How was this localization done, and why were two wires used?

After reviewing the specimen radiograph, what will you tell the surgeon?

This patient has pleomorphic calcifications in a segmental distribution extending for approximately 6 to 7 cm in the upper outer (craniocaudal view not shown) quadrant of the right breast. Two wires are used for the localization to bracket the location of the calcifications for the surgeon. To excise a lesion completely, the use of two or more wires is recommended in some patients when the lesion spans several centimeters or when dealing with multifocal or multicentric disease.

On the specimen radiograph, calcifications are noted extending to the edges of the specimen. This is discussed directly with the surgeon at the time of the surgery so that additional tissue can be obtained. Although the specimen is a two-dimensional representation of a three-dimensional structure, if the lesion of interest approximates one of the margins, this is discussed with the surgeon. With the patient still in the operating room, additional tissue can be taken in an effort to minimize the likelihood that a second operative procedure will be needed to obtain clear margins.

Following preoperative wire localizations, the specimen is placed in a container that allows for compression of the specimen with an alphanumeric grid. However, based on recent reports in the literature suggesting that vigorous specimen compression may result in “false positive” histologic interpretations of tumor extending to the margins, we apply only a minimal amount of compression. The specimen radiograph is done using magnification technique on a mammographic unit, or a dedicated specimen radiography unit. The use of an alphanumeric grid enables us to place a pin through the center of the lesion, or four pins can be used to delineate the margins of larger lesions for the pathologist. This assures that the area of mammographic concern is fully evaluated histologically.

The specimen radiograph is done to confirm that the localized lesion is excised, assess gross margin involvement, document removal of the localization wire, potentially identify additional unsuspected lesions, and mark the area of concern for the pathologist.

Because of the extent (6 to 7 cm) of the lesion in this patient, and the difficulty in obtaining clear margins with an acceptable resulting cosmetic effect, she went on to have a simple mastectomy. She also had a sentinel lymph node biopsy. Although sentinel lymph node biopsies are not done routinely in patients diagnosed with ductal carcinoma in situ (DCIS) on core biopsy, they are usually done in patients in whom an extensive area of DCIS is suspected based on the imaging findings. In these patients, the likelihood of microinvasive disease is increased and, because of the extent of the disease, it is possible that not all of the tissue will be evaluated histologically. In this patient, high-nuclear-grade ductal carcinoma in situ with central necrosis is diagnosed extensively involving the upper outer quadrant of the right breast. No invasion is identified in the tissue examined. The excised sentinel lymph nodes are normal [Tis, pN0(sn) (i>2), pMX, Stage 0].

 

 

Patient 34

Figure 4.34. Diagnostic evaluation, 55-year-old patient being evaluated for a cluster of calcifications, posteromedially in the right breast. Craniocaudal (A) view, right breast. Craniocaudal (B) view, photographically coned to the cluster of calcifications. Image (C) of the right breast using a fenestrated, alphanumeric spot compression paddle to determine the coordinates for the calcifications seen mammographically. Spot tangential (D) view of the metallic BB placed at the coordinates for the calcifications confirming that these are skin calcifications.

Although no lucent-centered calcifications are identified in the cluster, because of the close proximity to the skin, a skin location for these calcifications is suspected. What can you do to prove that these calcifications are on the skin and that biopsy is not needed?

If these calcifications are in the skin, a tangential view of the skin containing the calcifications should show that these are dermal. To obtain the tangential view, a “skin localization” is done. Craniocaudal and oblique views are reviewed to establish the shortest distance from the skin to the calcifications. In this patient, the calcifications are closest to the skin on the medial aspect of the breast, so a 90-degree mediolateral approach is taken. Normally, a regular, full compression paddle with an alphanumeric grid is used to compress the breast for localization. However, having a spot compression paddle with an alphanumeric grid is helpful in reaching lesions in hard-to-access locations including the axillary tail or anywhere posteriorly in the breast. The spot compression paddle facilitates the inclusion of tissue that may otherwise be difficult to include on an image with a full paddle.

A 90-degree mediolateral view using the spot compression paddle with an alphanumeric grid is obtained so that coordinates for the calcifications can be determined (Fig. 4.34C). The patient is maintained in compression until the coordinates for the calcifications are determined. A metallic BB is placed at C.5 and 2.5. Compression is released and a tangential view of the metallic BB is obtained. If the calcifications are in the skin, they will be imaged in tangent to the x-ray beam and in close association with the metallic BB (Fig. 4.34D). If they are not on the skin, they will be imaged in the breast parenchyma, not in tangent to the x-ray beam and at a distance from the metallic BB. In this patient, the calcifications are dermal in location, and further intervention or short-term follow-up is not indicated. Annual screening mammography is recommended.

BI-RADS® category 2: benign finding.

 

Bibliography

Agoff SN, Lawton TJ. Papillary lesions of the breast with and without atypical ductal hyperplasia. Am J Clin Pathol. 2004;122:440–443.

Asoglu O, Ugurlu MM, Blanchard K, et al. Risk factor for recurrence and death after primary surgical treatment of malignant phyllodes tumors. Ann Surg Oncol. 2004;11:1011–1017.

Berg WA. Image-guided breast biopsy and management of high-risk lesions. Radiol Clin N Am. 2004;42:935–946.

Berg WA, Mrose HE, Ioffe OB. Atypical lobular hyperplasia or lobular carcinoma in situ at core-needle biopsy. Radiology. 2001;218:503–509.

Brenner RJ, Jackman RJ, Parker SH, et al. Percutaneous core needle biopsy of radial scars of the breast: when is excision necessary? AJR Am J Roentgenol.2002;179:1179–1184.

Carder PJ, Garvican J, Haigh I, Liston JC. Needle core biopsy can reliably distinguish between benign and malignant papillary lesions of the breast. Histopathology. 2004;46:320–327.

Carder PJ, Murphy CE, Liston JC. Surgical excision is warranted following a core biopsy diagnosis of mucocele-like lesion of the breast. Histopathology. 2004;45:148–154.

Farshid G, Pieterse S, King JM, Robinson J. Mucocele-like lesions of the breast: a benign cause for indeterminate or suspicious mammographic microcalcifications. Breast J. 2005;11(1):15–22.

Fasih T, Jain M, Shrimankar J, et al. All radial scars/complex sclerosing lesions seen on breast screening mammograms should be excised. Eur J Surg Oncol. 2005;31:1125–1128.

Foster MC, Helvie MA, Gregory NE, et al. Lobular carcinoma in situ or atypical lobular hyperplasia at core needle biopsy: is excisional biopsy necessary? Radiology. 2004;231:813–819.

Gill HK, Ioffe OB, Berg WA. When is a diagnosis of sclerosing adenosis acceptable at core biopsy? Radiology. 2003;228:50–57.

Glazebrook K, Reynolds C. Mucocele-like tumors of the breast: mammographic and sonographic appearances. AJR Am J Roentgenol. 2003;180:949–954.

Greenstein-Orel S, Evers K, Yeh IT, et al. Radial scar with microcalcifications: radiologic-pathologic correlation. Radiology. 1992;183:479–482.

Guerra-Wallace MM, Chistensen WN, White RL. A retrospective study of columnar alteration with prominent apical snouts and secretions and the association with cancer. Am J Surg. 2004;188:395–398.

Günhan-Bilgen I, Memis A, Üstün EE, et al. Sclerosing adenosis: mammographic and ultrasonographic findings with clinical and histopathological correlation. Eur J Radiol. 2002;44:232–238.

Hamele-Bena D, Cranor ML, Rosen PP. Mammary mucocele-like lesions: benign and malignant. Am J Surg Pathol. 1996;20:1081–1085.

Ivan D, Selinko V, Sabin AA, et al. Accuracy of core needle biopsy diagnosis in assessing papillary breast lesions: histologic predictors of malignancy. Mod Pathol. 2004;17:165–171.

Jacobs TW, Byrne C, Colditz G, et al. Radial scars in benign breast biopsy specimens and the risk of breast cancer. N Engl J Med. 1999;340:430–436.

Jacobs TW, Chen YY, Guinee DG, et al. Fibroepithelial lesions with cellular stroma on breast core needle biopsy. Am J Clin Pathol. 2005;124:342–354.

Jacobs TW, Connolly JL, Schnitt SJ. Nonmalignant lesions in breast core needle biopsies. Am J Surg Pathol. 2002;26:1095–1110.

Jacobs TW, Natasha P, George K, Schnitt SJ. Carcinomas in situ of the breast with indeterminate features: role of E-cadherin staining in categorization. Am J Surg Pathol. 2001;25:229–236.

Kim JY, Han BK, Choe YH, Ko YH. Benign and malignant mucocele-like tumors of the breast: mammographic and sonographic appearances. AJR Am J Roentgenol. 2005;185:1310–1316.

Komenaka IK, El-Tmaer M, Pile-Spellman E, Hibschoosh H. Core needle biopsy as a diagnotic tool to differentiate phyllodes tumor from fibroadenoma. Arch Surg. 1003;138:987–990.

Liberman L. Clinical management issues in percutaneous core breast biopsy. Radiol Clin North Am. 2000;38:791–807.

Liberman L, Bracero N, Vuolo MA, et al. Percutaneous large-core biopsy of papillary breast lesions. AJR Am J Roentgenol. 1999;172:331–337.

Liberman L, Sama M, Susnik B, et al. Lobular carcinoma in situ at percutaneous breast biopsy: surgical biopsy findings. AJR Am J Roentgenol. 1999;173:291–299.

Mercado CL, Hamele-Bena D, Oken SM, et al. Papillary lesions of the breast at percutaneous core-needle biopsy. Radiology. 2006;238:801–808.

Page DL, Schuyler PA, Dupont WD, et al. Atypical lobular hyperplasia as a unilateral predictor of breast cancer risk: a retrospective cohort study. Lancet.2003;361:125–129.

Patterson JA, Scott M, Anderson N, Kirk SJ. Radial scar, complex sclerosing lesion and risk of breast cancer. Analysis of 75 cases in Northern Ireland. Eur J Surg Oncol. 2004;30:1065–1068.

Ramsaroop R, Greenberg D, Tracey N, Benson-Cooper D. Mucocele-like lesions of the breast: an audit of 2 years at Breast Screen Auckland (New Zealand). Breast J. 2005;11(5): 321–325.

Renshaw AA, Derhagopian RP, Tizol-Blanco DM, Gould EW. Papillomas and atypical papillomas in breast core needle biopsy specimens. Am J Clin Pathol. 2004;122:217–221.

Rosai J. Borderline epithelial lesions of the breast. Am J Surg Pathol. 1991;15:209–221.

Rosen EL, Bentley RC, Baker JA, et al. Image-guided core needle biopsy of papillary lesions. AJR Am J Roentgenol. 2002;179: 1185–1192.

Rosen, PP. Rosen's Breast Pathology. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2001.

Ung OA, Lee WB, Greenberg ML, Bilous M. Complex sclerosing lesion: the lesion is complex, the management is straightforward. ANZ J Surg. 2001;71:35–40.



If you find an error or have any questions, please email us at admin@doctorlib.info. Thank you!