What is the role of ultrasonography (US) in breast cancer screening?

Updated: Feb 01, 2017
  • Author: Paul R Fisher, MD; Chief Editor: Eugene C Lin, MD  more...
  • Print


Although mammography is an effective screening tool, data suggest that it is often less sensitive in detecting cancer in mammographically dense breast tissue. The use of US for screening for breast disease has not been generally recommended for high-risk women with dense breasts.

Although some research projects have reported reasonable results from US breast screening, a number of serious issues need to be solved before the practice is recommended for general application. Factors include interobserver variability, intraobserver variability, unknown sensitivity, and low specificity (leading to numerous biopsy evaluations of benign lesions). [10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29] Kolb et al and Buchberger et al found that, when performed carefully, ultrasonography may be useful in detecting occult breast cancer in dense breasts. [30, 31]

A retrospective study of 48,251 women who underwent  full-field digital mammography and ultrasound for breast cancer screening found that ultrasound alone is satisfactory for all age groups, although full-field digital mammography plus computer-aided detection plus ultrasound was found to be the perfect screening method. The detectability of breast cancer by ultrasound (96.5%) or full-field digital mammography plus computer-aided detection plus ultrasound  (100%) was superior to that of full-field digital mammography (87.1%) or full-field digital mammography plus computer-aided detection (88.3%). [1]

The American College of Radiology Imaging Newtwork 6666 study (ACRIN 6666) found that the cancer detection rate with ultrasound is comparable to that with mammography (58 of 111, vs 59 of 111, respectively), with a greater proportion of invasive cancers being detected by ultrasound than by mammography (91.4% vs 69.5%, respectively), but false positives were more common with ultrasound. The number of ultrasound screens to detect one cancer was 129, and for mammography, 127. [2]

US is generally acknowledged to be a highly operator dependent modality that requires a skilled practitioner, high-quality examinations, and state-of-the-art equipment. Currently, it is recommended that the use of US in screening for breast disease be reserved for special situations, such as for highly anxious patients who request it and for women who have a history of mammographically occult carcinoma.

In view of the results of these studies, a prospective, multicenter study was carried out to examine the role of US in breast cancer screening. A large multicenter study supported by the Avon Foundation and the National Institutes of Health was created through the American College of Radiology Imaging Network (ACRIN). [34] In this project, a protocol to assess the efficacy of screening breast US was implemented in 14 imaging centers to better define the role of US in breast cancer screening. The study reported higher cancer detection in high-risk women that underwent annual ultrasound screening in addition to mammography compared to those that underwent mammography alone. [35] (More information is available on the ACRIN Web site.) In September 2012, the U.S. Food and Drug Administration approved the first ultrasound system, the somo-v Automated Breast Ultrasound System (ABUS), for breast cancer screening in combination with standard mammographyspecificallyforwomenwith dense breast tissue. [36] ABUS is indicated for women with a negative mammogram, no breast cancer symptoms and no previous breast intervention such as surgery or biopsy.

See the images below of ultrasonography for breast cancer.

Breast cancer, ultrasonography. Mediolateral obliq Breast cancer, ultrasonography. Mediolateral oblique digital mammogram of the right breast in a 66-year-old woman with a new, opaque, irregular mass approximately 1 cm in diameter. The mass has spiculated margins in the middle third of the right breast at the 10-o'clock position. Image demonstrates both the spiculated mass (black arrow) and separate anterior focal asymmetry (white arrow).
Breast cancer, ultrasonography. This mediolateral Breast cancer, ultrasonography. This mediolateral mammogram was obtained in a 74-year-old woman with 2-week history of spontaneous discharge from the right nipple. A metal BB marker was placed on a possible lump at the 2-o'clock position. The breast is heterogeneously dense, which may decrease the sensitivity of mammography.
Breast cancer, ultrasonography. Craniocaudal scree Breast cancer, ultrasonography. Craniocaudal screening digital mammogram in a 46-year-old woman shows a new mass (arrow) at the 7- to 8-o'clock position in the right breast. Diagnostic mammography and sonography were then requested.
Breast cancer, ultrasonography. Radial sonogram sh Breast cancer, ultrasonography. Radial sonogram shows a mass that is nearly isoechoic relative to breast fat. The mass has angulated and spiculated margins surrounded by echogenic fibrous tissue. The margins are marked with white electronic calipers. Its largest dimension is 0.8 cm.
Breast cancer, ultrasonography. Digital spot compr Breast cancer, ultrasonography. Digital spot compression view of the left breast in a 79-year-old woman who presented with a palpable lump in the upper outer quadrant of the left breast. Image shows a BB marker over the palpable high-density mass, which is approximately 2 cm in diameter and has obscured margins.
Breast cancer, ultrasonography. This mediolateral Breast cancer, ultrasonography. This mediolateral oblique digital mammogram of the left breast was obtained in a 48-year-old woman with a several-month history of bloody discharge from the left nipple. Image demonstrates dilated ducts extending from the nipple into the lateral aspect of the breast (asterisks) with a calcification in 1 of the dilated ducts (arrowhead).
Breast cancer, ultrasonography. Color Doppler sono Breast cancer, ultrasonography. Color Doppler sonogram (displayed in black and white in the Doppler color box) from the same quadrant of the left breast demonstrates blood flow in the tumor within the ducts. The white oval areas (with central asterisks) represent blood flow within the intraductal tissue and thus confirms that the echogenic material within the ducts is tumor and not just intraluminal debris, blood clot, or secretions.
Breast cancer, ultrasonography. Spot magnification Breast cancer, ultrasonography. Spot magnification 90° mediolateral view of the mass in Image 33 demonstrates that it is heterogeneous, with a thin rim of subcapsular radiolucent fat (arrows).

A study of the positive predictive value (PPV) of bilateral whole-breast ultrasonography (BWBU) for detection of synchronous breast lesions on initial diagnosis of breast cancer found that BWBU can detect additional synchronous malignancy with a relatively high PPV, especially when mammography findings are correlated with ultrasound findings. In 75 patients who had synchronous lesions, PPV for additional biopsy was 25.7% (18 of 70). The PPV for synchronous lesions detected both on mammography and BWBU was 76.9% (10 of 13) and detected only on BWBU was 14.3% (7 of 49). A mass with calcification on mammography presentation (P< 0.01), presence of calcification among the ultrasonography findings (P< 0.01), and high Breast Imaging Reporting and Data System final assessment (P< 0.01) were imaging factors that were associated with malignancy in the additional synchronous lesion. [37]

In a retrospective study of women younger than 40 years identified with invasive cancer (N = 27) or ductal carcinoma in situ (N = 3), ultrasonography was found to be reliable as the primary imaging modality. Of the 30 women, 28 underwent mammography (graded as uncertain, suspicious, or malignant in the majority), and malignancy was missed in one patient. All 30 patients underwent ultrasonography (reported as uncertain, suspicious, or malignant, an indication for diagnostic core biopsy), and ultrasonography alone did not miss any cancers but did fail to detect multifocal disease in one patient. [38]

Did this answer your question?
Additional feedback? (Optional)
Thank you for your feedback!