Evaluation of Abnormal Mammography Results and Palpable Breast Abnormalities

Professional organizations recommend that women at the age of 40 or 50 years start undergoing screening mammography every 1 to 2 years (1-5). Primary care practitioners order most screening mammography examinations and must decide how to evaluate women who have an abnormal result. Evaluation of women with an abnormal mammography result is a common problem because even good-quality mammography facilities generally interpret 5% to 10% of all screening examinations as abnormal. About 90% of women with abnormal results do not have breast cancer (6-9); therefore, a safe and efficient evaluation is crucial. Breast symptoms are also a common problem; primary care practitioners receive approximately 20 presentations per 1000 person-years for the investigation of a breast symptom (10, 11). A breast lump is the most common symptom associated with breast cancer; between 9% and 11% of breast lumps result in a diagnosis of breast cancer (10, 12, 13). The prevalence of breast cancer among women who present with a breast lump increases with age from 1% for women 40 years of age and younger to 9% for women between 41 and 55 years of age to 37% for women age 55 years and older (12). Given that about 1 in 10 women with a breast lump or abnormal mammography result will have breast cancer, primary care practitioners must make a series of decisions to exclude or establish a diagnosis of breast cancer in these women. We review the literature on the evaluation of an abnormal screening mammography result and palpable breast abnormality and present an evidence-based approach with which to evaluate these two common problems. Methods We searched for published manuscripts determining the accuracy of screening and diagnostic mammography, fine-needle aspiration biopsy [FNAB], and core-needle biopsy to detect breast cancer among women with nonpalpable and palpable breast lesions. We systematically searched MEDLINE from January 1966 to March 2003 using the Medical Subject Heading terms or key words mammography, sensitivity, specificity, screening combined with breast cancer or breast neoplasm; or fine-needle biopsy, sensitivity, specificity, palpable mass combined with breast cancer or breast neoplasm; or core needle biopsy, sensitivity, specificity, palpable mass combined with breast cancer or breast neoplasm. We manually searched bibliographies of original and review articles identified in MEDLINE. Studies were included if 1) mammography, FNAB, or core-needle biopsy was performed before a definitive diagnosis was obtained; 2) the study sample included 100 or more women; and 3) breast cancer status was determined from histopathology review of excisional biopsy specimens, from linkage with a state cancer registry or the Surveillance, Epidemiology, and End Results program, or from clinical follow-up of 95% or more of the study sample. Studies of mammography, FNAB, or core-needle biopsy were considered high quality if they had a study sample that was population-based or consecutively sampled and if they determined cancer status in 95% or more of the study sample 1 or more years after the imaging or biopsy test was performed in order to determine the false-negative rate of the test (6, 9, 13-22). The major limitations of studies not considered high quality were incomplete follow-up, convenience study sample, and results that were not age adjusted. We primarily used results of three high-quality studies to describe optimal management strategies to evaluate nonpalpable and palpable breast lesions (13-15). Only one study (14) that evaluated the accuracy of screening mammography reported likelihood ratios or results to calculate likelihood ratios for the six American College of Radiology Breast Imaging and Reporting Data System (BI-RADS) assessment categories (23) for both first and subsequent screening examinations. A study of the accuracy of diagnostic mammography was the only population-based study to report results for nonpalpable and palpable breast lesions (15). The study of the accuracy of FNAB was the only study that gave results that allowed calculation of likelihood ratios for the four cytology assessment categories (13). Evaluation of Abnormal Screening Mammography Examination Screening mammography is performed in asymptomatic women with the goal of discovering invasive breast cancer at an early, curable stage. Screening mammography typically includes two views of each breast (craniocaudal and mediolateral oblique). The sensitivity of mammography ranges from 74% to 95%, and the specificity ranges from 89.4% to 99.1% (9, 14, 16, 17, 24). Sensitivity and specificity are higher for women more than 50 years of age, whereas sensitivity is lower and specificity higher for subsequent examinations compared with first screening examinations (9, 14, 16, 17, 24). The most common (and most worrisome) mammographic abnormalities that are found on screening examinations and that require further evaluation are masses and calcifications. The differential diagnosis for a mammographic mass includes cyst, benign nonproliferative lesions, benign proliferative lesions with or without atypia, fibroadenoma, radial scar, intramammary lymph node, lipoma, galactoceles, ductal carcinoma in situ, and invasive cancer. The differential diagnosis for a mammographic calcification includes benign nonproliferative lesions, benign proliferative lesions with or without atypia, fat necrosis, atherosclerosis, dermal lesion, ductal carcinoma in situ, and invasive cancer. No specific mammographic findings are associated with lobular carcinoma in situ. However, when mammographic calcifications are biopsied, lobular carcinoma in situ has been identified adjacent to histologic calcifications located in normal epithelium. Of note, although the differential diagnosis for a mass or calcification is long, all diagnoses other than ductal carcinoma in situ and invasive cancer are benign and require no further evaluation. Radiologists generally describe both masses and calcifications in terms of location, size, and other characteristics (such as shape, borders, and pattern). In addition to describing findings, radiologists make an assessment and recommendation (25). The American College of Radiology recommends one of six assessments for interpretation of a mammographic screening examination (Tables 1 and 2) (23). Abnormal screening mammography assessments are evaluated with diagnostic mammography, ultrasonography, and biopsy. Table 1. American College of Radiology Breast Imaging Reporting and Data System Table 2. Frequency of Screening Mammographic Results and Risk for Breast Cancer Based on Mammographic Result Diagnostic Mammography and Ultrasonography of Nonpalpable Lesions Diagnostic mammography is a comprehensive radiologic examination of a breast abnormality that may allow the radiologist to more definitively classify a finding; it can sometimes be done during the same visit as the screening examination. Diagnostic mammography consists of multiple specialized views, including magnification views or spot compression views. Results of diagnostic mammography are reported by using one of five assessment categories by the American College of Radiology: negative, benign finding, probably benign finding, suspicious abnormality, or highly suggestive of malignancy (23). About 15% of women with nonpalpable cancer will have a diagnostic mammography examination that shows no evidence of cancer (Table 3) (15). Negative likelihood ratios for a normal mammography result in women with a nonpalpable lesion are approximately 0.2 (Table 3). Ultrasonography is of particular value in distinguishing a cyst from a solid lesion on screening or diagnostic mammography. Ultrasonography is 98% to 100% accurate in diagnosing simple cysts when four rigorous sonographic criteria are used to evaluate the lesion (oval or lobulated shape; anechoic, well-defined posterior border; increased through-transmission; and no alteration of surrounding breast parenchyma) (26, 27). Table 3. Accuracy of Diagnostic Mammography in Women with a Nonpalpable Lesion or Breast Lump Biopsy For mammographic abnormalities that are nonpalpable and that require biopsy, image-guided tissue sampling is necessary. Tissue for diagnosis can be obtained by mammography, ultrasonography-guided FNAB, or core-needle biopsy or by open surgical biopsy with needle localization. The sensitivity of needle-localized excisional biopsy is 99% for nonpalpable lesions (28). The most common reason for missing the carcinoma is erroneous placement of the needle guidewire. Image-guided FNAB or core-needle biopsy is quicker, cheaper, and easier than standard-needle localization open biopsy (29-31). For image-guided FNAB or core-needle biopsy, a mammographic x-ray tube is angled to produce two views of the lesion, and the position of the lesion is calculated from the apparent movement of the lesion relative to a fixed reference grid. With use of this information, a needle is placed in the lesion; needle position is confirmed on repeated stereotactic x-ray views. Alternatively, high-resolution ultrasonography can be used to position the needle in real time within the lesion. A fine needle (22- to 25-gauge), an automated core needle (14- to 18-gauge), or a vacuum-assisted biopsy probe (11-gauge) can be used to obtain a sample of breast tissue. With nonpalpable lesions, core-needle biopsy is usually preferred over FNAB because the core sample provides adequate tissue for histologic diagnosis and is more accurate (32, 33). The accuracy of FNAB for nonpalpable lesions varies more than that of core-needle biopsy (sensitivity, 77% to 97%; specificity, 78% to 98%) (32-34); FNAB is also highly operator dependent and more often produces insufficient diagnostic material (33% for FNAB vs. 1.5% for core-needle biopsy) (18, 19). Image-guided FNAB allows safe sampling of very thin breast tissue and of lesions situated close to the chest wall or the skin, where core biopsy is not