Clinical Perspective
FOCUS ON: Women's Imaging
January 23, 2015

Screening for Dense Breasts: Digital Breast Tomosynthesis


OBJECTIVE. Digital breast tomosynthesis (DBT) is a recent imaging technology that was developed to address the limitations of conventional 2D mammography. The limitations of standard mammography are well known and include reduced sensitivity in dense breasts. Clinical research studies of DBT and the implementation of DBT have revealed that DBT has potential benefits for evaluating patients with dense breasts. This article will discuss the benefits and limitations of DBT as a screening alternative for women with dense breasts.
CONCLUSION. Studies to date have revealed that the use of DBT reduces recall rates and increases cancer detection rates. This has been demonstrated with the use of DBT for both screening and diagnostic purposes, as well as with imaging dense breasts. DBT has the ability to reduce breast tissue overlap, thus potentially revealing lesions that would otherwise have been missed. The limitations of DBT include longer interpretation times, higher costs, and increased radiation dose. These limitations present challenges that radiologists must consider before DBT implementation.
The relationship between breast tissue density and breast cancer risk was described as early as 1976 when Wolfe [1] reported his early findings. Since then, the significance of breast density as a risk factor has been debated. It has been reported that there is a two- to sixfold increase in the risk of breast cancer in women with dense breast tissue compared with those with a fatty pattern [2]. This increase in risk is significant because 50% of women between 40 and 49 years old and 30% of women from 70 to 79 years old have at least 50% dense breasts [3]. Studies have shown the attributable risk for breast cancer increases with greater breast density [4, 5]. In 1995, Byrne et al. [5] reported an almost five-fold increased risk of breast cancer in women who had a breast density of 75% or greater as compared with women who had fatty breasts. To put this increased risk related to breast density in perspective, fewer than 5% of breast cancers are attributable to mutations of the breast cancer gene (BRCA1 and BRCA2) [6]. Furthermore, it has been suggested [7] that mammography screening effectiveness is reduced in patients with high-risk parenchymal patterns (i.e., heterogeneously dense and extremely dense).

The Dense Tissue Debate

The American College of Radiology's fifth edition of the BI-RADS [8] has recently changed the designation of breast composition. The recent description changes came during a time in breast imaging when attention was being paid to the breast density debate. The BI-RADS change was made to eliminate the percentages used for each density category and rather to highlight the description associated with each category; for example, when the breasts are heterogeneously dense, it is noted that this pattern may obscure small masses.
Roubidoux et al. [9] discussed the implications of dense breast tissue on cancer detection using mammography; they reported that the association between large tumor size and dense tissue may be due to either more rapidly growing tumors in breasts with increased glandular tissue or increased difficulty in tumor detection on mammography. The latter is a well-known limitation of conventional 2D mammography technology and is, in part, the reason we continue to investigate additional screening alternatives to evaluate patients with dense breast tissue.
In the past several years, legislation about breast density imaging findings has made its way through the United States, with 19 states to date having a breast density notification law. The intent of the law is to notify patients who have dense breast tissue that they may benefit from additional screening studies. Breast ultrasound is most commonly used for screening because of its wide availability and lack of ionizing radiation. Screening breast ultrasound does have some limitations such as a high false-positive rate and interobserver variability [10]. Breast MRI may be used for screening purposes in specific high-risk patients; however, the examination takes a long time and requires expensive equipment and additional staff, and image interpretation is time-consuming for the radiologist. Additionally, implementing MRI screening of patients with dense breasts would be extremely costly compared with options such as screening ultrasound and digital breast tomosynthesis (DBT). Identifying additional tools for screening women with dense breasts is an ongoing effort. DBT is a new technology that has been shown to have a role for breast cancer screening in women with dense breast tissue.


A known limitation of conventional 2D mammography is tissue superimposition, which can contribute to breast lesions being obscured. Tissue superimposition can occur to a greater degree in dense breast tissue; it has been reported that only half of cancers will be visible in extremely dense breast tissue [9]. As technology has advanced, the sensitivity of mammography has improved. In the Digital Mammographic Imaging Screening Trial, Pisano et al. [11] reported the accuracy of full-field digital mammography (FFDM) to be significantly higher than that of film-screen mammography in women with heterogeneously dense and extremely dense breasts. DBT is a newer breast imaging tool that is gaining widespread adoption because it has been shown to improve on the limitations of 2D mammography, particularly improving on the detection of lesions in dense breast tissue.
The DBT device is an FFDM system capable of producing standard 2D images and tomosynthesis images [12]. The 2D and DBT datasets can be acquired independently or in combination under the same compression. For image acquisition, the x-ray tube moves in a limited arc across the breast and a series of low-dose images are acquired from different angles. The images are then reconstructed into 1-mm slices, although a slice can be as thin as 0.5 mm. The ability to view the breast in slices can provide improved lesion visibility within the cross section of the breast tissue. Early in the clinical implementation of DBT, the results of studies promised that DBT would help to improve the accuracy of diagnostic and screening mammography, reduce recall rates, and provide accurate 3D lesion localization [1220].
The DBT imaging method is hoped to help to improve the detection and characterization of breast lesions especially in women with nonfatty breasts [12]. In a recently published article in Anticancer Research, Mariscotti and colleagues [13] described their study of 200 newly diagnosed breast cancer patients who had undergone digital mammography, DBT, ultrasound, and MRI. The aim of the study was to define the accuracy of the diagnostic tests and to calculate the sensitivity according to breast density. The study reported improved sensitivity with DBT in comparison with digital mammography (90.7% vs 85.2%, respectively). A 97.7% sensitivity was reported with combined digital mammography, DBT, and ultrasound. The addition of MRI brought sensitivity up to 98.8%. When specifically evaluating sensitivity based on breast density, Mariscotti and colleagues found that DBT maintained good sensitivity in dense breasts. The sensitivity of ultrasound was better than digital mammography and DBT for detecting lesions in dense breasts, and MRI maintained the highest sensitivity in all breast density types.
The thin-slice display of DBT should allow superior detection of lesions; the primary benefit of DBT is expected to be for the detection of noncalcified findings such as masses, architectural distortions, and asymmetries [12]. Additional studies have shown comparable detection of calcifications for DBT and FFDM [14, 20]. In our clinical experience, we found equal or higher conspicuity of microcalcifications when comparing DBT with FFDM [14]. Increased lesion conspicuity is greatly beneficial when imaging dense breast tissue because it is can be difficult to perceive suspicious lesions in extremely dense breast tissue. The border of masses, number of masses, and associated findings such as dilated ducts or vessels around a mass are better depicted on DBT images, especially in dense breasts [15]. Because of the ability of DBT to reduce tissue superimposition, a benefit of DBT was a reduction in the recall rate in women with dense breasts. Haas and colleagues [16] reported that the addition of DBT reduced recall rates for all breast density groups and age groups, with significant differences in recall rates for scattered heterogeneously dense and extremely dense breasts; these results ultimately led them to conclude that the greatest reductions in recall rates were in women younger than 50 years old and in women with dense breasts [16]. Their study findings reiterate the belief that DBT will prove to be beneficial for patients with dense breast tissue and for those with nondense breast tissue.
Rose et al. [17] reported in 2013 the results of an observational study comparing recall rates, biopsy rates, cancer detection rates, and positive predictive values for radiologists with and without the use of DBT. The results of that study showed improvements with the use of DBT in all performance measures evaluated and that these improvements were distributed across all breast density categories. These findings indicate that DBT can contribute to increased cancer detection and reduced recall rates in patients with dense breasts and in those with fatty or scattered tissue.
Rafferty and colleagues [18] reported their findings after implementing DBT. When using DBT with FFDM for imaging patients with dense breasts, a proportionally greater advantage was seen when comparing AUCs: 0.877 for DBT with FFDM, 0.786 for FFDM alone, and 0.832 for FFDM with craniocaudal DBT. Margolies and colleagues [19] evaluated a cohort of patients at increased risk of breast cancer and investigated the effect of adding DBT on patient management. They found that patients with BI-RADS category 3 or 4 breast density had a significantly higher likelihood of having DBT findings that changed management (13% of patients with heterogeneously dense or extremely dense tissue vs 9% of patients with fatty or scattered tissue; p = 0.03). However, even in that study, DBT changed management in patients with the less dense breast patterns also. When evaluating cancer detection, Margo-lies and colleagues ultimately state that their data did indicate that DBT was most beneficial for patients with increased breast density, given that the three additional cancers detected on DBT were found in patients with heterogeneous or extremely dense breasts.
An additional role for DBT is as a method for the estimation of breast density. The benefit of measuring breast density with DBT is that the entire breast volume is included. The use of DBT for estimating breast density can be beneficial because, in addition to more accurately measuring breast density, DBT will aid in more accurate risk assessment with regard to breast density. To date, several different methods have been used to assess breast density, such as the classification proposed by Wolfe [1] and semiquantitative and quantitative computer-aided techniques [21]. Breast density estimation using current mammographic techniques can be limited by the 2D nature of the technology, and DBT can potentially improve these breast density assessments [21]. Early work on this topic has shown that automated breast density estimation eliminates subjectivity between 2D mammography and DBT and is more reproducible than nonautomated techniques. Tagliafico and colleagues [21] evaluated this issue further by performing a study to compare breast density estimates from 2D mammography with those from DBT according to different BI-RADS categories using automated software. This issue—how breast density estimation may differ between 2D mammography and DBT for the different breast density categories—has not been studied to date in the literature. They found that breast density values were significantly lower on DBT than on 2D FFDM but that density values showed a nonlinear relationship across BI-RADS categories. Higher differences between breast density estimations for 2D FFDM and DBT were observed for the least dense and the most dense breasts.
Radiation dose is important to consider when thinking about using DBT as a screening tool, given that we need to be mindful of dose management in the screening environment. Early work on DBT dose [22] has shown that combination imaging of a homogeneous representation of a 5-cm-thick 50% glandular breast can result in a dose lower than the 3-mGy limit proposed in the Mammography Quality Standards Act. Furthermore, these investigators found that DBT acquisition for some thicker, denser breasts (75% and 100% glandular) resulted in a lower mean glandular dose, finding it to be as low as 67% as much radiation as an FFDM acquisition [22]. The authors note that this higher dose for FFDM was the result of a very high tube current–exposure time product being chosen by the automatic exposure control setting.
Since the implementation of DBT, researchers have been striving to find a method to reduce dose. In May of 2013, the use of software that creates a synthetic image, which is called the “C-View” (Hologic) on the Selenia Dimensions 3D DBT system (Hologic), received U.S. Food and Drug Administration (FDA) approval. The C-View image is a summation image of the tomosynthesis slices acquired for a view that eliminates the need for the FFDM portion of the combination FFDM and DBT examination. This capability is important because using C-View will reduce dose given that only DBT is required while allowing an overview of the DBT slices and providing an image for comparison with prior FFDM studies.
The results of an initial study of C-View revealed some limitations [23]: Investigators reported decreased sensitivity with comparable specificity with the use of the synthetic view and DBT in comparison with the use of FFDM and DBT. The researchers concluded that improved synthesized images would be needed to eliminate double exposure during DBT screening [23]. However, in a recently published article in Radiology, Skaane et al. [24] reported that C-View and DBT performed comparably to FFDM and DBT when looking at cancer detection rates. These results are indicative of improvements to the synthesized images and possibly to improved reader confidence in the interpretation and the appearance of the C-View images.
The interpretation time for DBT is an additional factor that will need to be considered in high-volume screening facilities when deciding whether to use DBT as a screening modality. Most studies report reading times that are doubled in comparison with 2D FFDM reading times. Dang et al. [25] studied the effect of implementing screening tomosynthesis on reading times. They evaluated 1502 combination 2D FFDM and DBT studies and reported that 23.8 studies were interpreted per hour with a 2.8-minute mean interpretation time per study. The interpretation of 2163 FFDM-only studies resulted in 34.0 studies per hour with a mean interpretation time of 1.9 minutes per study. Skaane and colleagues [26] also recently reported on increased interpretation times: 45 seconds for FFDM alone and 91 seconds for combination FFDM and DBT. Interpretation time is an important factor to consider when transitioning to DBT, especially early in the transition process. The increased reading times for each study will impact patient scheduling, radiologists' time, and a practice's daily workflow, especially in the screening environment. The increased reading time is important to be aware of when considering using the technology as an additional tool to screen patients with dense breast tissue. With experience, interpretation times should improve but will always remain longer than FFDM because of the multiple DBT slices that must be viewed.

Clinical Experience Using Digital Breast Tomosynthesis

Our facility has been gradually transitioning to DBT since FDA approval in 2011. We currently have six units capable of DBT imaging, with the remaining units at this time performing FFDM. Because we are working in a hybrid environment (both DBT and FFDM), we have had to design an imaging protocol to determine which patients will undergo DBT. Dense breast tissue, among several other factors, is a factor we have chosen to direct patients to DBT imaging.
Our facility reviewed our early experience with DBT implementation by evaluating a group of patients who underwent FFDM alone and a group who underwent combination DBT and FFDM [27]. Of the patients in the combination DBT and FFDM group, 61.64% had dense breast tissue compared with 54.20% in the FFDM-alone group. We found that the combination DBT plus FFDM group had a significantly lower recall rate than the FFDM-alone group despite having additional risk factors including dense breast tissue.
Many facilities are operating in a hybrid environment, whereas others have been able to transition to DBT. An article recently published in JAMA [28] shows the benefits a facility can have from implementing DBT, whether it is a complete conversion or a hybrid conversion. The study was retrospective and reviewed two breast cancer screening periods; period 1 was FFDM only and period 2 was FFDM plus DBT. An important fact about this study was that it included 13 academic and nonacademic breast centers, some that had converted totally to DBT and some with a hybrid conversion. This study looked at recall rates for additional imaging, cancer detection rates, and positive predictive values for recall and for biopsy. The study evaluated 454,850 examinations (281,187 FFDM and 173,663 FFDM and DBT) and found that the recall rate dropped with the use of FFDM and DBT compared with FFDM alone. The biopsy rate increased and the cancer detection rate (and specifically the invasive breast cancer detection rate) increased with FFDM and DBT compared with FFDM alone. The carcinoma in situ rate was the same at 1.4 per 1000 examinations for both periods. Adding DBT was associated with increases in the positive predictive values for recall and for biopsy and a decrease in recall rates. It would be interesting to look at these rates across the breast density categories to assess the performance of DBT in patients with higher breast densities.


DBT is a new imaging modality that has potential for screening women with dense breasts. In addition, studies to date have revealed that the use of this technology reduces recall rates and increases cancer detection rates and that DBT has value in the diagnostic arena. DBT has the ability to reduce breast tissue overlap, thus potentially revealing lesions that would have otherwise been missed. The limitations of DBT include longer interpretation times, higher costs, and increased radiation dose. These limitations present challenges that radiologists must consider before DBT implementation.


Wolfe JN. Risk for breast cancer development determined by mammographic parenchymal pattern. Cancer 1976; 37:2486–2492
Porter GJ, Evans AJ, Cornford EJ, et al. Influence of mammographic parenchymal pattern in screening-detected and interval invasive breast cancers on pathologic features, mammographic features, and patient survival. AJR 2007; 188:676–683
Stomper PC, D'Souza DJ, DiNitto PA. Analysis of parenchymal density on mammograms in 1353 women 25–79 years old. AJR 1996; 167:1261–1265
Boyd NF, Byng JW, Jong RA, et al. Quantitative classification of mammographic densities and breast cancer risk: results from the Canadian National Breast Screening Study. J Natl Cancer Inst 1995; 87:670–675
Byrne C, Schairer C, Wolfe J, et al. Mammographic features and breast cancer risk: effects with time, age, and menopause status. J Natl Cancer Inst 1995; 87:1622–1629
Harvey JA, Bovbjerg VE. Quantitative assessment of mammographic breast density: relationship with breast cancer risk. Radiology 2004; 230:29–41
Sala E, Warren R, McCann J, Duffy S, Day N, Luben R. Mammographic parenchymal patterns and mode of detection: implications for the breast screening programme. J Med Screen 1998; 5:207–212
Sickles EA, D'Orsi CJ, Bassett LW, et al. ACR BIRADS mammography. In: ACR BI-RADS Atlas, Breast Imaging Reporting and Data System. Reston, VA: American College of Radiology, 2013
Roubidoux MA, Bailey JE, Wray LA, Helvie MA. Invasive cancers detected after breast cancer screening yielded a negative result: relationship of mammographic density to tumor prognostic factors. Radiology 2004; 230:42–48
Berg WA, Blume JD, Cormack JB, et al. Combined screening with ultrasound and mammography vs mammography alone in women at elevated risk of breast cancer. JAMA 2008; 299:2151–2163
Pisano ED, Gatsonis C, Hendrick E, et al.; Digital Mammographic Imaging Screening Trial (DMIST) Investigators Group. Diagnostic performance of digital versus film mammography for breast-cancer screening. N Engl J Med 2005; 353:1773–1783
Helvie MA. Digital mammography imaging: breast tomosynthesis and advanced applications. Radiol Clin North Am 2010; 48:917–929
Mariscotti G, Houssami N, Durando M, et al. Accuracy of mammography, digital breast tomosynthesis, ultrasound and MR imaging in preoperative assessment of breast cancer. Anticancer Res 2014; 34:1219–1225
Destounis S, Arieno A, Morgan R. Preliminary clinical experience with digital breast tomosynthesis in the visualization of breast microcalcifications. J Clin Imaging Sci 2013; 3:65
Park JM, Franken EA Jr, Garg M, Fajardo LL, Niklason LT. Breast tomosynthesis: present considerations and future applications. RadioGraphics 2007; 27(suppl 1):S231–S240
Haas BM, Kalra V, Geisel J, Raghu M, Durand M, Philpotts LE. Comparison of tomosynthesis plus digital mammography and digital mammography alone for breast cancer screening. Radiology 2013; 269:694–700
Rose SL, Tidwell AL, Bujnoch LJ, Kushwaha AC, Nordmann AS, Sexton R. Implementation of breast tomosynthesis in a routine screening practice: an observational study. AJR 2013; 200:1401–1408
Rafferty EA, Park JM, Philpotts LE, et al. Assessing radiologist performance using combined digital mammography and breast tomosynthesis compared with digital mammography alone: results of a multicenter, multireader trial. Radiology 2013; 266:104–113
Margolies L, Cohen A, Sonnenblick E, et al. Digital breast tomosynthesis changes management in patients seen at a tertiary care breast center. ISRN Radiol 2014; 2014:658–929
Kopans D, Gavenonis S, Halpern E, Moore R. Calcifications in the breast and digital breast tomosynthesis. Breast J 2011; 17:638–644
Tagliafico AS, Tagliafico G, Cavagnetto F, Calabrese M, Houssami N. Estimation of percentage breast tissue density: comparison between digital mammography (2D full field digital mammography) and digital breast tomosynthesis according to different BI-RADS categories. Br J Radiol 2013; 86:20130255. Epub 2013 Sept 12
Feng SS, Sechopoulos I. Clinical digital breast tomosynthesis system: dosimetric characterization. Radiology 2012; 263:35–42
Gur D, Zuley ML, Anello MI, et al. Dose reduction in digital breast tomosynthesis (DBT) screening using synthetically reconstructed projection images: an observer performance study. Acad Radiol 2012; 19:166–171
Skaane P, Bandos AI, Eben EB, et al. Two-view digital breast tomosynthesis screening with synthetically reconstructed projection images: comparison with digital breast tomosynthesis with full-field digital mammographic images. Radiology 2014; 271:655–663
Dang PA, Freer PE, Humphrey KL, Halpern EF, Rafferty EA. Addition of tomosynthesis to conventional digital mammography: effect on image interpretation time of screening examinations. Radiology 2014; 270:49–56
Skaane P, Bandos AI, Gullien R, et al. Comparison of digital mammography alone and digital mammography plus tomosynthesis in a population-based screening program. Radiology 2013; 267:47–56
Destounis S, Arieno A, Morgan R. Initial experience with combination digital breast tomosynthesis plus full field digital mammography or full field digital mammography alone in the screening environment. J Clin Imaging Sci 2014; 4:9
Friedewald SM, Rafferty EA, Rose SL, et al. Breast cancer screening using tomosynthesis in combination with digital mammography. JAMA 2014; 311:2499–2507

Information & Authors


Published In

American Journal of Roentgenology
Pages: 261 - 264
PubMed: 25615747


Submitted: August 1, 2014
Accepted: September 15, 2014


  1. breast density
  2. breast imaging
  3. digital breast tomosynthesis
  4. mammography
  5. screening



Stamatia V. Destounis
All authors: Elizabeth Wende Breast Care, LLC, 170 Sawgrass Dr, Rochester, NY 14620.
Renee Morgan
All authors: Elizabeth Wende Breast Care, LLC, 170 Sawgrass Dr, Rochester, NY 14620.
Andrea Arieno
All authors: Elizabeth Wende Breast Care, LLC, 170 Sawgrass Dr, Rochester, NY 14620.


Address correspondence to S. V. Destounis ([email protected]).

Metrics & Citations



Export Citations

To download the citation to this article, select your reference manager software.

Articles citing this article

View Options

View options


View PDF

PDF Download

Download PDF







Copy the content Link

Share on social media