Emily L. Sedgwick

The Breast Ultrasound Lexicon: Breast Imaging Reporting and Data System (BI-RADS)

o Aradiologist’s ability to perceive an abnormality on an image is not unlike an art critic’s ability to perceive the ignificance of color choice when evaluating a painting. Pereption is a learned skill, refined over time, and difficult to uantify. Some have better perception than others. Radiology esidents are given the example of “Aunt Minnie,” a term oined by the late Dr Edward B. Neuhauser, the former chief f Radiology at Boston Children’s Hospital. An “Aunt Minnie” efers to a radiology finding pathognomonic for a disease.1 A radiologist should be able to perceive a finding because he or she has seen it many times before, like one recognizes his or her Aunt Minnie. Unfortunately, not all imaging findings are pathognomonic for a specific disease. Consequently, radiologists need to identify several different features in an attempt to specifically characterize a lesion. Breast cancer can present on imaging studies in many fashions, requiring a combination of perceptions to make a specific diagnosis. Mammography remains the gold standard for the detection of breast cancer, although the sensitivity is approximately 75%.2 Ultrasound has been used primarily as a complementary tool to mammography to discern solid masses from cysts, thereby improving the specificity of mammography.3 Sonography is an perator-dependent modality. When operator-dependence s combined with the variety of ways a feature can be decribed, the utility of ultrasound is hampered. Stavros et al3 defined several ultrasound characteristics to predict the likelihood of malignancy when evaluating a solid mass. Other authors attempted to reproduce the results of Stavros et al with varied success.4,5 Consequently, the creation of a lexicon was advocated for better communication of findings and so that standard terms could be used as the basis for ultrasound research.4 The bulk of breast ultrasound research pertains to he use of sonography in conjunction with mammography.

BI-RADS update for breast cancer caregivers.

This review will discuss changes relevant to breast cancer caregivers in the fifth edition of the Breast Imaging Reporting and Data System.

Granulomatous Mastitis in a Transgender Patient

Granulomatous mastitis is a rare and benign inflammatory condition of the breast most commonly affecting women of child-bearing age as well as patients on oral contraceptives. This condition is important to identify due to its diagnostic mimicry of malicious entities such as breast carcinoma. Clinical and radiological findings are nonspecific and may overlap with breast carcinomas, thus pathologic confirmation is often necessary for definitive diagnosis. Although cases of granulomatous mastitis have been described in cisgender females, there have been no reported cases in the transgender patient, a growing patient population with few imaging guidelines. Transgender patients are at risk of developing this breast entity due to the use of long-term hormone treatments or presence of residual breast tissue. A trial of antibiotics or steroids may be administered. However, surgical treatment is often necessary in recurrent or refractory cases.

Is breast magnetic resonance imaging (MRI) useful for diagnosis of additional sites of disease in patients recently diagnosed with pure ductal carcinoma in situ (DCIS)

PURPOSE To determine if breast MRI is useful for detecting additional or invasive sites of disease in patients initially diagnosed with pure DCIS. MATERIALS AND METHODS A retrospective review of women diagnosed with pure DCIS who underwent a breast MRI for evaluation of extent of disease was performed at a single institution from January 2013 to April 2015. Data analysis included imaging (mammography, ultrasound and MRI) and pathology characteristics (histology and biomarker status) of the primary DCIS as well as descriptors for the additional sites of disease incidentally found by breast MRI. RESULTS A total of 108 patients were diagnosed with pure DCIS during this time period. A breast MRI for staging was recommended for all patients. 76 patients had an MRI performed, ages ranging from 38 to 79 years old (median, 53 years); sizes ranging from 0.3 to 10cm (mean, 2.2cm). A total of 52 patients (68%) either had suspicious new finding(s) (n=27, 36%) or bigger tumor size than originally visualized on mammography (n=43, 57%). A total of twenty-seven patients (36%) had other MRI findings suspicious for additional sites of disease in either breast (four in the ipsilateral breast and twenty-three in the contralateral breast). From this group of patients, twenty-three (85%) patients underwent MRI-guided biopsy as recommended. The four patients who did not have the recommended MRI guided-biopsy either underwent total mastectomies or refused the biopsy. Six out of the twenty-three patients (26%) were diagnosed with an additional site of cancer (5 DCIS and 1 IDC) (7.9%, CI=3.7%, 16.2%). All of the six patients had contralateral disease (100%) and none had a second site of disease in the ipsilateral breast. The size of the additional sites of disease ranged from 0.4 to 8cm (mean, 3.1cm) and the size of the primary lesion in this selected group ranged from 0.1 to 10.9cm (mean, 5.6cm). Ages ranged from 44 to 63 years old (median, 52.5 years). Five out 6 patients (83%) presented with the first site of disease as pure DCIS with estrogen (ER) and progesterone (PR) receptors positive and one case (17%) was pure DCIS ER/PR- negative. The second incidental lesion found on MRI demonstrated 5 cases of contralateral pure DCIS and 1 case of contralateral invasive disease. From this group, we did not have the data for biomarker analysis for the second site of disease in 2 cases and 3 cases showed concordant biomarker status between the first and second sites of disease. The 1 case that presented with an invasive component in the contralateral side of the initially biopsy-proven pure DCIS had discordant biomarkers compared to the first site of disease: the first site of pure DCIS was ER/PR-negative and the second site of invasive ductal carcinoma (IDC) presented with ER/PR-positive status. CONCLUSION From a total of 76 patients with recent diagnosis of pure DCIS who underwent staging breast MRI examination for diagnosis of additional sites of disease, approximately 8% (95% confidence interval=3.7%, 16.2%) was diagnosed with an additional site of cancer and 1.3% (95% confidence interval=0.2%, 7%) of the total cases had invasive disease in the additional sites with different biomarker status; changing their management and prognosis.

Stereotactic Biopsy of Segmental Breast Calcifications: Is Sampling of Anterior and Posterior Components Necessary?

RATIONALE AND OBJECTIVES Core needle biopsy results of segmental calcifications on mammography can have direct impact on surgical management. Although dependent on breast size, cancer spanning greater than 5 cm is usually treated with mastectomy, and cancer less than 5 cm is managed with lumpectomy. Approach to stereotactic biopsy of morphologically similar segmental calcifications that span more than 5 cm on mammography varies geographically and is currently largely based on preference of the surgical or medical oncology colleagues. Some clinicians prefer biopsy of the anterior and posterior aspects of the abnormality, whereas others believe a single biopsy within the abnormality is adequate. There is insufficient data to support whether a single biopsy of calcifications is adequate to establish the need for mastectomy, or if pathology-proven cancer in the anterior and posterior components to define the extent of disease is required. This study aims to evaluate concordance rates of paired biopsies of suspicious segmental mammographic calcifications. MATERIALS AND METHODS From a 5-year review of our imaging database, 32 subjects were identified with breast imaging reporting and data system (BI-RADS) 4 or 5 segmental calcifications on mammography who underwent anterior and posterior stereotactic biopsies. The paired biopsy results were independently analyzed for concordance on benign, high-risk, or malignant pathology. RESULTS Of the 32 cases, there was perfect agreement (32/32 cases = 100% concordance, 95% confidence interval = 89.3-100%) in anterior and posterior pairs in benign, high-risk, or malignant findings (kappa = 1, P < 0.001). CONCLUSIONS The absence of data on pathological concordance in anterior and posterior aspects of suspicious, morphologically similar, segmental calcifications spanning 5 cm or more has led to a varied clinical approach to stereotactic biopsy. The 100% rate of pathological concordance in our study suggests that a single biopsy is adequate for diagnosis and representative of the whole mammographic abnormality. Implementation of this approach will potentially reduce unnecessary biopsies and surgeries, minimize healthcare costs, and decrease patient morbidity.

Practical Considerations for Integrating Digital Breast Tomosynthesis Into Clinical Practice

DESCRIPTION OF THE PROBLEM Mammography was introduced in the 1960s for the early detection of breast cancer. The main mode of acquisition was film screen until the development of digital detectors in the 1990s, which made digital mammography possible. Mammography, however, has received criticism for the “overdiagnosis” of lesions of uncertain clinical significance and the induction of patient anxiety due to false positives, which most commonly occur in women with dense breasts. Digital breast tomosynthesis (DBT) overcomes these limitations of conventional mammography by decreasing the impact of noise from superimposition of normal breast structures. To date, many studies have shown that DBT increases mammography’s sensitivity and specificity. It increases cancer detection because of increased lesion conspicuity when breast structure noise is reduced [1-4]. We are a seven-person academic radiology breast imaging department in a publically funded health system (SmithClinic ofHarrisHealth System, Houston, Texas). The Smith Clinic is the main hub for diagnostic breast imaging services provided to Harris Health System patients. Ten satellite facilities provide 2-D screening mammography. In October 2012, three Selenia Dimensions units (Hologic,

Spindle cell lesions of the breast: Multimodality imaging and clinical differentiation of pathologically similar neoplasms

Spindle cell lesions of the breast comprise a wide-range of entities including reactive, benign and malignant proliferations. They can be pathologically challenging to differentiate as there is often immunohistochemical and morphologic similarities with characteristic spindle shaped cellular patterns. Radiological and pathological correlation is essential. Radiology detects, defines the size and extent, and assists in localizing the lesions. Pathology confirms the diagnosis and provides prognostic parameters. Familiarity with the clinicoradiological features of these diagnostically challenging lesions helps to establish an accurate pathological diagnosis and subsequent clinical decision making.