See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/312930463

BI-RADS® fifth edition: A summary of changes

Article in Diagnostic and Interventional Imaging · January 2017

DOI: 10.1016/j.diii.2017.01.001

CITATIONS READS

408 39,224

5 authors, including:

Basak Erguvan Dogan

University of Texas Southwestern Medical Center
166 PUBLICATIONS 3,862 CITATIONS

SEE PROFILE

All content following this page was uploaded by Basak Erguvan Dogan on 28 August 2018.

The user has requested enhancement of the downloaded file.

 Diagnostic and Interventional Imaging (2017) 98, 179—190

REVIEW /Breast Imaging

®
BI-RADS fifth edition: A summary of
changes
D.A. Spak a,∗, J.S. Plaxco b,c, L. Santiago b,
M.J. Dryden b, B.E. Dogan b,d

a
McGovern Medical School, Department of Diagnostic and Interventional Radiology, MSB
2.130B, 6431 Fannin Street, Houston, TX 77030, USA
b
The University of Texas MD Anderson Cancer Center, Division of Diagnostic Imaging, 1515
Holcombe Blvd., Unit 1350, Houston, TX 77030, USA
c
Radiology of Huntsville, 2006 Franklin Street, Suite 200, Huntsville, AL 35801, USA
d
University of Texas Southwestern Medical Center, Department of Radiology, 2201 Inwood
Drive, Dallas, TX 75390, USA

KEYWORDS ®
Abstract The Breast Imaging Reporting and Data System (BI-RADS ) is a standardized system of
Breast imaging; reporting breast pathology as seen on mammogram, ultrasound, and magnetic resonance imag-
BI-RADS; ing. It encourages consistency between reports and facilitates clear communication between
Lexicon; the radiologist and other physicians by providing a lexicon of descriptors, a reporting structure
Mammography; that relates assessment categories to management recommendations, and a framework for data
Magnetic resonance ®
collection and auditing. This article highlights the changes made to the BI-RADS atlas 5th edi-
imaging (MRI) tion by comparison with its predecessor, provide a useful resource for a radiologist attempting
to review the recent changes to the new edition, and serve as a quick reference to those who
have previously become familiar with the material.
© 2017 Editions françaises de radiologie. Published by Elsevier Masson SAS. All rights reserved.

®
The Breast Imaging Reporting and Data System (BI-RADS ) is a standardized system of
reporting breast pathology encountered on mammography, ultrasound, and magnetic reso-
nance imaging (MRI). This structured system encourages consistency between reports and
facilitates clear communication between the radiologist and other physicians by provid-
ing a lexicon of descriptors, a reporting structure that relates assessment categories to
management recommendations, and a framework for data collection and auditing [1—3].
®
Published by the American College of Radiology (ACR), the initial edition of BI-RADS was
created in 1993 [1]. The second, third, and fourth editions were released in 1995, 1998,

∗ Corresponding author.
E-mail addresses: [email protected], [email protected] (D.A. Spak).

http://dx.doi.org/10.1016/j.diii.2017.01.001
2211-5684/© 2017 Editions françaises de radiologie. Published by Elsevier Masson SAS. All rights reserved.

Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
 180 D.A. Spak et al.

and 2003 respectively [2]. Each edition was built on previ- The purpose of this paper was to summarize the changes
®
ous work by clarifying previous terms with an aim toward introduced within the newest edition of BI-RADS . It is the
risk stratification. Furthermore, the inclusion of ultrasound hope of the authors that this review will provide a useful
and MRI descriptors promoted congruence across modalities. resource for a radiologist attempting to review the recent
Artist renderings of pathology, which appeared in the third changes to the new edition and serve as a quick reference to
edition of the lexicon also assisted learning and understand- those who have previously become familiar with the mate-
ing the framework. rial.
The hiatus between the 4th and 5th lexicon allowed
several questions pertaining to the reproducibility of assess-
ment and frequency of certain descriptor use to be
®
Mammography
evaluated and published. The 5th edition of BI-RADS
released in 2013 ushered in numerous changes geared Breast composition
toward these goals. These changes include multiple new
descriptors added in recognition of the increasing risk of Breast composition or density is a comparison of the relative
malignancy an image finding represents, historic descriptors amounts of fat versus fibroglandular tissue in the breast. In
removed due to underutilization or redundant applica- the 4th edition of lexicon, composition was associated with
tion, and name changes to pre-existing descriptors to align percent values of fibroglandular tissue as assessed by the
descriptions across the three imaging modalities [1,2]. More- radiologist.
over, newer technology such as shear-wave elastography The terms available to describe breast composition
has been incorporated into the descriptors. Finally, images on a mammogram no longer include percentage quartiles
acquired directly from mammography, ultrasound, and MRI (Table 1). Radiologists’ assessment of numeric quartiles has
replace artist rendering to demonstrate pathology in the 5th consistently been mismatched with automated volume cal-
edition. culations particularly as fibroglandular volume increases [4].

®
Table 1 Summary of changes within mammography between the BI-RADS 4th and 5th editions [3].
4th edition 5th edition
Breast composition Previously based on percentages Percentages removed:
(A) Almost entirely fat (less than 25%) (A) The breasts are almost entirely fatty
(B) Scattered fibroglandular densities (B) There are scattered areas of
(25—50%) fibroglandular density
(C) Heterogeneously dense (50—75%) (C) The breasts are heterogeneously
(D) Extremely dense (greater than 75%) dense, which may obscure small masses
(D) The breasts are extremely dense,
which lowers the sensitivity of
mammography
Masses Historically masses: shape: lobular Removed in the 5th edition
Calcifications: typically benign: rim Historically rim, eggshell or Renamed to calcifications: typically
and lucent-centered lucent-centered calcifications benign: rim
calcifications
Calcifications: suspicious Historic categories removed: Now a single category calcifications:
morphology intermediate concern, suspicious suspicious morphology
calcifications and higher probability
malignancy
Asymmetries Historically special cases: global Asymmetries:
asymmetry and focal asymmetry (A) Asymmetry
(B) Global asymmetry
(C) Focal asymmetry
(D) Developing asymmetry
Intramammary lymph node and Historically located within special cases New separate categories
solitary dilated duct categories
Skin lesion Historically located within associated New separate category
features
Location of lesion: distance from Historically only depth was available to New in 5th edition
the nipple describe distance from the nipple
Use of radiopaque markers Definition of use expanded
Assessment categories 3, 4, and 5 Assigned after diagnostic workup
completed (not appropriate for
screening examinations)

Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
 ®
BI-RADS fifth edition: A summary of changes 181

The numerical values previously associated with them were ‘‘microlobulated’’, ‘‘indistinct’’, and ‘‘spiculated’’ may be
also removed to avoid confusion with the overall assessment used to describe margins as applicable.
categories. The descriptions within the 5th edition range
from least to most dense, and are based purely on the radi- Typically benign calcifications
ologist’s visual assessment of the amount of fibroglandular
®
density, again based on rough quartile distribution (Fig. 1; In the new BI-RADS 5th edition lexicon, typically benign
Table 1). calcification morphology includes ‘‘skin’’, ‘‘vascular’’,
‘‘coarse or popcorn-like’’, ‘‘large rod-like’’, ‘‘round’’,
Masses ‘‘rim’’, ‘‘dystrophic’’, ‘‘milk of calcium’’, and ‘‘suture’’.
®
The historic descriptors ‘‘eggshell’’ and ‘‘lucent-centered’’
Mass shape classification in the BI-RADS 5th edition con- have been removed and consolidated into the single calcifi-
sists of ‘‘oval’’, ‘‘round’’, and ‘‘irregular’’. The historic cation morphology descriptor of ‘‘rim’’ to describe calcium
shape descriptor of ‘‘lobular’’ has been removed from the surrounding a lucent center. The aim of this change was
lexicon. This is due to the possible confusion that could to simplify reporting as all three appearances are typically
arise from the term ‘‘lobular’’ applied to define the shape benign.
of a mass which normally conveys a benign etiology and
the margin descriptor ‘‘microlobulated’’ which implies a Calcifications with suspicious morphology
suspicious finding. Per the new lexicon, a mass with two
or three undulations may be described as ‘‘oval’’ and In the 4th edition, calcification morphology was examined in
the margin descriptors of ‘‘circumscribed’’, ‘‘obscured’’, three main subcategories: benign, intermediate and suspi-
cious. In the 5th edition, suspicious category morphology
replaces two previous categories ‘‘intermediate concern,
suspicious calcifications’’ and ‘‘higher probability malig-
nancy’’.
®
In the new BI-RADS 5th edition lexicon, the typically
suspicious calcification morphology includes ‘‘amorphous’’,
‘‘coarse heterogeneous’’, ‘‘fine pleomorphic’’, and ‘‘fine
linear or fine-linear branching’’. The change was brought
about after multiple studies concluded the chance of malig-
nancy was high enough to warrant biopsy [1,5—8]. Currently
‘‘amorphous’’, ‘‘coarse heterogeneous’’, and ‘‘fine pleo-
morphic’’ calcifications are thought to represent a lower
risk of malignancy and should be assigned to category 4B as
opposed to ‘‘fine linear or fine-linear branching’’ calcifica-
tions which should be assigned to category 4C (Fig. 2).

Asymmetries
Asymmetries are usually an artifact created from the sum-
®
mation of overlying tissue. The previous BI-RADS special
cases descriptors ‘‘global asymmetry’’ and ‘‘focal asym-
metry’’ have been significantly expanded in the current
edition. A new category for asymmetries has been created
with the following descriptors now available: ‘‘asymmetry’’,

Figure 1. Craniocaudal (CC) views from screening mammogram

obtained in different women. (a) 54-year-old woman with almost
entirely fatty breast. (b) 55-year-old woman with scattered areas of Figure 2. A 56-year-old woman undergoing screening mammo-
fibroglandular density. (c) 57-year-old woman with heterogeneously gram. Zoomed craniocaudal view of the right breast depicts fine
dense breast. (d) 50-year-old woman with extremely dense breast linear calcifications (arrow). Stereotactic breast biopsy revealed
composition. ductal carcinoma in situ, with high nuclear grade.

Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
 182 D.A. Spak et al.

Solitary dilated duct

A solitary dilated duct was previously described under
the ‘‘special cases’’ category in the 4th edition (Fig. 4).
Although rare, a ‘‘new findings’’ category has been created
in the current lexicon due to the increasing risk of malig-
nancy the finding represents [10].

Descriptors for location of lesion

The report of lesion location is encouraged to follow this
syntax: laterality (referring to the left or right breast); quad-
rant and clock face; depth (anterior, middle, or posterior);
and, distance from the nipple. This expanded description of
lesion location assists comparing findings across modalities.
Note that retroareolar refers to a central lesion within the
anterior depth.

Use of radiopaque markers

The use of radiopaque markers has been expanded upon
in the 5th edition. The use now discusses application of
markers for skin and palpable lesions. The 5th edition also
recommends each facility adopt a policy governing the type
of maker used for each lesion and a method of detailing what
the marker represents as an image annotation or description
within the final report.
®
Mammography BI-RADS assessment categories
®
Some clarification on the appropriate use of BI-RADS 3
assessment category has been undertaken in the 5th edition.
Specifically, in the context of a screening-detected abnor-
®
mality, BI-RADS assessment category 3 should be reserved
Figure 3. A 42-year-old woman undergoing diagnostic mammo-
for lesions that have been fully evaluated in the diagnostic
gram with personal history of lobular carcinoma in situ in the right setting and are inappropriate for screening examinations.
breast. (a) Right craniocaudal (CC) view; (b) right mediolateral Similarly, assessment categories 4 and 5 should also be
oblique (MLO) view; (c) left CC view; and (d) left MLO view demon- reserved until after the diagnostic workup of a lesion has
strate focal asymmetry within the right breast lower outer quadrant been completed.
(arrows). Follow-up magnetic resonance-guided biopsy yielded lob-
ular carcinoma in situ and pseudoangiomatous stromal hyperplasia.

‘‘global asymmetry’’, ‘‘developing asymmetry’’, and ‘‘focal

asymmetry’’ (Fig. 3). The ‘‘developing asymmetry’’ descrip-
tor was added to the lexicon as this finding has been
associated with a risk of malignancy high enough to warrant
recall and potential biopsy [9].

Intramammary lymph nodes

In the 4th edition of the lexicon, a descriptor for an intra-
mammary lymph node resided within the ‘‘special cases’’
category. A new separate category for ‘‘intramammary
lymph node’’ has been created in the 5th edition.

Skin lesions
Previously in the 4th edition, the descriptor ‘‘skin lesion’’
resided within the ‘‘associated features’’ category. In the
current lexicon a separate category has been created. These Figure 4. A 58-year-old woman undergoing screening mammo-
lesions should be marked by the mammography technologist gram without personal history of cancer. (a) Left craniocaudal
if apparent at time of examination. view and (b) left mediolateral oblique view demonstrate a solitary,
dilated duct (arrows) stable for several years.

Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
 ®
BI-RADS fifth edition: A summary of changes 183

Ultrasound aligning findings identified on separate modalities. Fur-

thermore, it helps to eliminate confusion of benign and
General considerations malignant mass descriptors across multiple imaging types.
To this end, the historic descriptor of ‘‘lesion boundary’’
A general consideration subsection has been added to the has been removed from the new version of the lexicon.
®
5th edition of BI-RADS . This subsection includes informa- This is in part based on the low diagnostic value of the
tion involving breast anatomy, image quality, labeling and presence of an echogenic transition zone or echogenic rim
measurement, as well as documentation. versus an abrupt margin transition zone of a mass to discrim-
inate benign and malignant pathology. The lexicon provides
Masses margin descriptors of ‘‘circumscribed’’, ‘‘indistinct’’,
‘‘angular’’, ‘‘microlobulated’’, and ‘‘spiculated’’ for mass
®
The 5th edition of the BI-RADS provides mass descriptors margins identified sonographically.
examined under ultrasound of ‘‘shape’’, ‘‘orientation’’, Finally, the historic descriptor of ‘‘posterior acoustic fea-
‘‘margin’’, ‘‘echo pattern’’, and ‘‘posterior features’’ tures’’ has been renamed in the new edition to ‘‘posterior
(Table 2). The unification of mass descriptors across mam- features’’ as mentioned above. The terms available to
mography and magnetic resonance examinations assists in describe these features of a mass in the 5th edition are ‘‘no

®
Table 2 Summary of changes within ultraound between the BI-RADS 4th and 5th editions [3].
4th edition 5th edition
General considerations subsection N. S. (A) Breast anatomy
(B) Image quality
(C) Labeling and measurement
(D) Documentation
Lesion boundary of masses Lesion boundary Lesion boundary removed from 5th
edition
Posterior features of masses Historically termed: posterior acoustic New term: posterior features
features
Macrocalcifications Historically under calcifications Removed from 5th edition
Intraductal calcifications N. S. Created under calcifications
Surrounding tissue attributes now Surrounding tissue: Associated features:
found under associated features (A) Architectural distortion (A) Architectural distortion
(B) Duct changes (B) Duct changes
(C) Edema (C) Edema
(D) Skin thickening (D) Skin thickening
(E) Skin retraction (E) Skin retraction
Cooper’s ligament changes Historically under surrounding tissue Now described under architectural
distortion
Vascularity under associated Historically vascularity included: Vascularity lexicon now under
features (A) Not present or not assessed associated features:
(B) Present in lesion (A) Absent
(C) Present immediately adjacent to (B) Internal vascularity
lesion (C) Vessels in rim
(D) Diffusely increased vascularity in
surrounding tissue
Elasticity assessment N. S. Created under associated features:
elasticity assessment:
(A) Soft
(B) Intermediate
(C) Hard
Special cases N. S. Special cases created:
(A) Simple cyst
(B) Postsurgical fluid collection
(C) Fat necrosis
Vascular abnormalities N. S. Special cases: vascular abnormalities:
(A) Arteriovenous
malformation/pseudoaneurysms
(B) Mondor disease
N.S.: information not previously specified.

Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
 184 D.A. Spak et al.

Elasticity assessment
The evaluation of tissue stiffness has been a developing area
of research related to breast cancer to augment morphologic
description. Several studies have demonstrated promis-
ing results in supplementing an ultrasound examination
with compression elasticity imaging to help differenti-
ate between benign and malignant lesions [11—13]. New
descriptors under the associated features category for elas-
ticity assessment of ‘‘soft’’, ‘‘intermediate’’, and ‘‘hard’’
have been created to further this goal.

Descriptors for special cases

Descriptors under the special cases category remain avail-
able from the previous edition to describe a unique diagnosis
or finding. These terms include: ‘‘clustered microcysts’’,
Figure 5. A 77-year-old woman with remote history of invasive
‘‘complicated cyst’’, ‘‘mass in or on skin’’, ‘‘foreign body
and in situ ductal carcinoma. Cropped grayscale ultrasound image
including implants’’, ‘‘lymph nodes intramammary’’, and
depicts dilated ducts extending from the nipple with intraductal cal-
cifications (arrows). Ultrasound-guided vacuum-assisted core biopsy ‘‘lymph nodes axillary’’. Note that a complicated cyst refers
of the calcifications yielded high-grade ductal carcinoma in situ with to a cyst with internal debris visible as homogeneous internal
comedonecrosis. low level echoes.
New descriptors are now available under special cases
in the 5th edition. The terms ‘‘simple cyst’’, ‘‘postsurgical
posterior features’’, ‘‘enhancement’’, ‘‘shadowing’’, and fluid collection’’, and ‘‘fat necrosis’’ have been created
‘‘combined pattern’’. as well as descriptors related to vascular abnormalities
of ‘‘arteriovenous malformations (AVM)/pseudoaneurysms’’
Calcifications and ‘‘Mondor disease’’ which can be used to describe these
findings with a pathognomonic diagnosis (Fig. 7).
The lexicon added ‘‘intraductal calcifications’’ in
the 5th edition and removed the historic descriptor
‘‘macrocalcifications’’ (Fig. 5). ‘‘Calcifications in a mass’’ MRI
and ‘‘calcifications outside of a mass’’ are also available
to describe calcifications visible during an ultrasound Amount of fibroglandular tissue
examination.
The amount of fibroglandular tissue is now described
in a similar manner as within mammography. The new
Reorganization of surrounding tissue
descriptors
Multiple terms historically located within the ‘‘surrounding
tissue’’ category have been reorganized under ‘‘associated
features’’. These descriptors include ‘‘architectural dis-
tortion’’, ‘‘duct changes’’, ‘‘edema’’, ‘‘skin retraction’’
and ‘‘skin thickening’’. Note that ‘‘architectural distor-
tion’’ describes various effects upon surrounding tissue
including changes involving Cooper’s ligaments which was
historically located under the surrounding tissue category
(Fig. 6). This change aligns the lexicon in a similar manner
to the descriptors available within mammography. However,
these findings are currently described in MRI as non-mass
features.

Vascularity
Terms related to vascularity were historically ‘‘not present
or not assessed’’, ‘‘present in lesion’’, ‘‘present imme-
diately adjacent to lesion’’, and ‘‘diffusely increased
vascularity in surrounding tissue’’. Findings related to vas- Figure 6. A 57-year-old woman with family history of breast can-
cer and developing asymmetry on recent mammogram. Cropped
cularity have now been streamlined within the 5th edition
grayscale ultrasound image demonstrates an area of architec-
under the associated features category as being ‘‘absent’’, tural distortion (arrow). Follow-up ultrasound-guided core biopsy
‘‘internal vascularity’’, and ‘‘vessels in rim’’. revealed invasive lobular carcinoma.

Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
 ®
BI-RADS fifth edition: A summary of changes 185

the fibroglandular tissue should be evaluated on the first

post-contrast image and may be described as ‘‘minimal’’,
‘‘mild’’, ‘‘moderate’’, or ‘‘marked’’.
‘‘Symmetric’’ and ‘‘asymmetric’’ descriptors of non-
mass-like enhancement are also found within the category of
background parenchymal enhancement. These terms have
historically been located under a separate finding of non-
mass enhancement.

Focus descriptor
Foci has been removed from the historic descriptor
focus-foci as new descriptors of background parenchy-
mal enhancement detailed previously are now available.
A focus of enhancement should have no correlating
Figure 7. A 59-year-old woman with recently diagnosed ductal findings on pre-contrast images before applying this descrip-
carcinoma in situ of the right breast. (a) Cropped grayscale ultra- tor.
sound image demonstrates a dilated non-compressible periareolar
superficial vein (arrow); (b) Cropped grayscale ultrasound image
with power Doppler applied demonstrates decreased internal flow
compatible with superficial thrombophlebitis or Mondor disease.
Mass shape
A mass identified during a magnetic resonance examination
descriptors of ‘‘almost entirely fat’’, ‘‘scattered fibrog- can be given an ‘‘oval’’, ‘‘round’’, or ‘‘irregular’’ shape
landular tissue’’, ‘‘heterogeneous fibroglandular tissue’’, descriptor. The historic ‘‘lobular’’ descriptor was removed
and ‘‘extreme fibroglandular tissue’’ are now employed to as it served a redundant role with the ‘‘oval’’ descriptor
describe breast composition (Fig. 8; Table 3). which includes lobulated masses. This change is among many
to the descriptors to characterize a breast mass that were
Background parenchymal enhancement implemented to align them with the terminology utilized in
other modalities.
New terms to quantify the amount of background parenchy-
mal enhancement present on contrast enhanced MRI studies Mass margin
are available. Quantifying the level of enhancement is
important as the sensitivity of detecting malignancy has The margin of a mass may be described as ‘‘circumscribed’’,
found to decrease as the level of background parenchymal ‘‘irregular’’, or ‘‘spiculated’’. The historic ‘‘smooth’’
enhancement increases [14]. The level of enhancement of descriptor has been renamed to ‘‘circumscribed’’.

Figure 8. T1-weighted magnetic resonance images obtained in the transverse plane in four different women. (a) A 77-year-old woman
with known breast cancer receiving an MRI for staging purpose. T1-weighted magnetic resonance image shows fibroglandular tissue with
almost entirely fat; (b) a 37-year-old woman at high risk for breast cancer receiving a screening MRI. T1-weighted magnetic resonance image
shows scattered fibroglandular tissue; (c) a 35-year-old woman obtaining MRI due to right nipple discharge and lack of ultrasound correlate.
T1-weighted magnetic resonance image demonstrates heterogeneous fibroglandular tissue; (d) a 59-year-old woman receiving a screening
MRI. T1-weighted magnetic resonance image shows extremely dense fibroglandular tissue.

Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.

186
®
Table 3 Summary of changes within MRI between the BI-RADS 4th and 5th editions [3].
4th edition 5th edition
Breast composition N. S. Amount of fibroglandular tissue
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.

(A) Almost entirely fat

(B) Scattered fibroglandular tissue
(C) Heterogeneous fibroglandular tissue
(D) Extreme fibroglandular tissue
Background parenchymal enhancement N. S. (A) Minimal
(B) Mild
(C) Moderate
(D) Marked
Non-mass-like enhancement: symmetric or Historically under non-mass-like enhancement Now under background parenchymal enhancement
asymmetric
Focus Historically focus-foci Foci (of focus-foci) removed
Shape: lobular Previous mass descriptor Removed
Margin: circumscribed Historically margin: smooth Renamed to margin: circumscribed
Margin: not circumscribed: irregular and Historically under mass: margin Now under mass: margin: not circumscribed
spiculated (A) Irregular
(B) Spiculated
Internal enhancement characteristics Historically under mass: mass enhancement Now under masses: internal enhancement
(A) Homogeneous characteristics
(B) Heterogeneous enhancement (A) Homogeneous
(C) Rim enhancement (B) Heterogeneous enhancement
(D) Dark internal septations (C) Rim enhancement
(D) Dark internal septations
Removal of enhancing internal septation and Historically under mass: mass enhancement Removed
central enhancement (A) Enhancing internal septation
(B) Central enhancement
Removal of non-mass-like enhancement terms Historically under non-mass-like enhancement Removed
(A) Distribution modifiers: ductal
(B) Internal enhancement: stippled, punctate
(C) Internal enhancement: reticular, dendritic
Clustered ring enhancement pattern N. S. Non-mass enhancement: internal enhancement
patterns: clustered ring
New finding categories N. S. New finding categories
(A) Intramammary lymph node
(B) Skin lesion
New terms under associated features N. S. Associated features
(A) Skin invasion: direct invasion

D.A. Spak et al.

(B) Skin invasion: inflammatory cancer
(C) Axillary adenopathy
(D) Chest wall invasion
(E) Architectural distortion
Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.

BI-RADS fifth edition: A summary of changes

Table 3 (Continued)
4th edition 5th edition

®
Renamed terms under associated features Historically: nipple retraction or inversion Renamed: nipple retraction
Removal of terms under associated features N. A. Removed from associated features
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.

(A) Pre-contrast high duct signal

(B) Edema
(C) Lymphadenopathy
(D) Hematoma-blood
(E) Abnormal signal void
(F) Cyst
Addition of fat-containing lesions N. S. Fat-containing lesions
(A) Lymph nodes: normal
(B) Lymph nodes: abnormal
(C) Fat necrosis
(D) Hamartoma
(E) Postoperative seroma, hematoma with fat
Kinetic curve assessment: initial phase Historically Now
(A) Slow (A) Slow
(B) Medium (B) Medium
(C) Rapid (C) Fast
Implant material and lumen type N. S. Implants: implant material and lumen type
(A) Saline
(B) Silicone: intact
(C) Silicone: ruptured
(D) Other implant materials
(E) Lumen type
Implant location N. S. Implants: implant location
(A) Retroglandular
(B) Retropectoral
Implant focal bulge N. S. Implants: abnormal implant contour: focal bulge
Intracapsular silicone findings N. S. Implants: intracapsular silicone findings
(A) Radial folds
(B) Subcapsular line
(C) Keyhole sign (teardrop, noose)
(D) Linguine sign
Extracapsular silicone N. S. Implants: extracapsular silicone
(A) Breast
(B) Lymph nodes
Implants: water droplets N. S. Implants: water droplets
Implants: peri-implant fluid N. S. Implants: peri-implant fluid
N. S.: information not previously specified; N. A.: not applicable.

187
 188 D.A. Spak et al.

Mass internal enhancement characteristics Skin lesions

The descriptors of ‘‘homogeneous’’, ‘‘heterogeneous’’, The separate category of skin lesion has been created within
‘‘rim enhancements’’, and ‘‘dark internal septations’’ have the 5th edition. Benign enhancing pathology along the skin
been relocated under the internal enhancement char- such as enhancing keloids or dermatitis are appropriate for
acteristics category which have previously been located this category.
under the historic category mass enhancement. The historic
descriptors of ‘‘enhancing internal septation’’ and ‘‘central New associated features descriptors
enhancement’’ have been removed due to underutilization.
Many of the descriptors have carried over from the 4th
edition including ‘‘nipple retraction’’, ‘‘nipple invasion’’,
Non-mass enhancement distribution ‘‘skin retraction’’, and ‘‘pectoralis muscle invasion’’. New
® descriptors now found under associated features include
The 5th edition of BI-RADS details multiple descriptors for
‘‘axillary adenopathy’’, ‘‘architectural distortion’’, and
the distribution of non-mass enhancement which should be
‘‘chest wall invasion’’ (Fig. 10). ‘‘Skin invasion’’ has also
applied to describe areas of enhancement without convex
been added with delineation between ‘‘direct invasion’’ and
borders. The descriptors available to describe the distribu-
‘‘inflammatory cancer’’ (Fig. 11).
tion include ‘‘focal’’, ‘‘linear’’, ‘‘segmental’’, ‘‘regional’’,
‘‘multiple regions’’, and ‘‘diffuse’’. The historic descriptor
‘‘ductal’’ has been removed due to underutilization.
Removed associated features descriptors
Several terms within the associated features category for
Non-mass enhancement internal enhancement MRI examinations were removed because of overlap or
patterns due to the creation of a new category. Examples of these
changes include ‘‘pre-contrast high duct signal’’, ‘‘edema’’,
To characterize the internal enhancement patterns of ‘‘lymphadenopathy’’, ‘‘hematoma-blood’’, ‘‘abnormal sig-
non-mass enhancement the lexicon provides the descrip- nal void’’, and ‘‘cyst’’.
tors ‘‘homogenous’’, ‘‘heterogeneous’’, ‘‘clumped’’, and
‘‘clustered ring’’. The addition of ‘‘clustered ring’’ was Fat-containing lesions
added to describe suspicious enhancement of periductal
stroma (Fig. 9). The historic term ‘‘reticular, dendritic’’ was Several terms to denote fat-containing pathology were
removed due to underutilization as well as ‘‘stippled, punc- added into the lexicon. Descriptors for these lesions include
tate’’ as this pattern can be characterized by descriptors ‘‘fat necrosis’’, ‘‘postoperative seroma/hematoma with
within background parenchymal enhancement. fat’’, and ‘‘hamartoma’’ (Fig. 12). ‘‘Lymph nodes’’ may
also be described as fat-containing if adipose tissue is
demonstrated with further delineation as ‘‘normal’’ or
Intramammary lymph nodes ‘‘abnormal’’.

The separate category of intramammary lymph node has Kinetic curve assessment
been created with the 5th edition of the lexicon.
The assessment of the kinetic curve is separated between
the initial phase and the delayed phase. The initial

Figure 9. A 35-year-old woman with inflammatory breast can-

cer presenting for restaging after systemic therapy. T1-weighted Figure 10. A 63-year-old woman with invasive lobular carci-
fat-suppressed MRI obtained in the transverse plane following noma of the left breast. T1-weighted fat-suppressed MR image in
administration of a gadolinium chelate demonstrates marked left the transverse plane after administration of a gadolinium chelate
breast skin thickening and clustered ring enhancement (arrow). demonstrates chest wall invasion as enhancement of the underlying
Subsequent MRI- guided biopsy revealed granulomatous mastitis. chest wall and intercostal space (arrow).

Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
 ®
BI-RADS fifth edition: A summary of changes 189

Figure 11. A 42-year-old woman with history of inflammatory right breast cancer, invasive ductal carcinoma, and ductal carcinoma in situ.
T1-weighted fat-suppressed MR image in the transverse plane obtained after intravenous administration of a gadolinium chelate depicts
marked global asymmetric skin thickening due to inflammatory breast cancer (arrow).

phase of the kinetic curve may be described as ‘‘slow’’, Miscellaneous

‘‘medium’’, or ‘‘fast’’ which was historically titled
‘‘rapid’’. The descriptors available for the delayed phase Implant locations. ‘‘Retroglandular’’ and ‘‘retropectoral’’
remain ‘‘persistent’’, ‘‘plateau’’, and ‘‘washout’’. descriptors may be used to describe breast implant location.
Abnormal implant contour. The descriptor ‘‘focal bulge’’
has been created under abnormal implant contour to
Implant material and lumen type describe this form of implant irregularity.
Intracapsular silicone findings. Terms relating to radial
Terms describing numerous facets of breast augmentation folds and several common radiology signs associated
are available within the 5th edition of the lexicon. To with breast implants are provided in the new edition
describe breast material and lumen type the descriptors of the lexicon. The descriptors include ‘‘radial folds’’,
‘‘saline’’, ‘‘silicone: intact’’, ‘‘silicone: ruptured’’, ‘‘other ‘‘subcapsular line’’, ‘‘keyhole sign’’, and the ‘‘linguine
implant material’’ and ‘‘lumen type’’ are now available. sign’’ (Fig. 13).
The lumen type may also be characterized as ‘‘single’’, Extracapsular silicone. ‘‘Lymph nodes’’ and ‘‘breast’’
‘‘double’’, or ‘‘other’’. Example alternative materials used are now available to describe the location of extra-
in breast implants are soy oil, polypropylene, polyurethane, capsular silicone observed on a magnetic resonance
and sponges. examination.

Figure 12. A 31-year-old woman undergoing MR examination for palpable abnormality without mammographic or sonographic correlate.
(a) T1-weighted MR image obtained in the transverse plane demonstrates a fat-containing mass unrelated to the palpable abnormality
compatible with known hamartoma (arrow); (b) T2-weighted fat-suppressed MR image obtained in the sagittal plane of the hamartoma
demonstrates the interwoven fat within the mass (arrow); (c) cropped mediolateral oblique mammogram of the hamartoma (arrow).

Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
 190 D.A. Spak et al.

[2] Burnside ES, Sickles EA, Bassett LW, Rubin DL, Lee CH, Ikeda
DM, et al. The ACR BI-RADS experience: learning from history.
J Am Coll Radiol 2009;6:851—60.
[3] American College of Radiology. ACR BI-RADS 5th edition
changes. http://www.acr.org/∼/media/acr/documents/pdf/
qualitysafety/resources/birads/birads v5 changes.pdf
[accessed on 01.01.17].
[4] Morrish OWE, Tucker L, Black R, Willsher P, Duffy SW, Gilbert
FJ. Mammographic breast density: comparison of methods for
quantitative evaluation. Radiology 2015;275:356—65.
[5] Liberman L, Abramson AF, Squires FB, Glassman JR, Morris
EA, Dershaw DD. The Breast Imaging Reporting and Data Sys-
tem: positive predictive value of mammographic features and
final assessment categories. AJR Am J Roentgenol 1998;171:
35—40.
[6] Berg WA, Arnoldus CL, Teferra E, Bhargavan M. Biopsy of amor-
phous breast calcifications: pathologic outcome and yield at
Figure 13. MR images obtained in two different women with stereotactic biopsy. Radiology 2001;221:495—503.
breast implants. (a) A 65-year-old woman with history of right breast [7] Burnside ES, Ochsner JE, Fowler KJ, Fine JP, Salkowski LR,
cancer following mastectomy and implant placement. T2-weighted Rubin DL, et al. Use of microcalcification descriptors in BI-
MR image obtained in the sagittal plane demonstrates the linguine RADS 4th edition to stratify risk of malignancy. Radiology
sign associated with intracapsular implant rupture (arrow); (b) a 2007;242:388—95.
74-year-old woman with a history of breast cancer following mas- [8] Bent CK, Bassett LW, D’Orsi CJ, Sayre JW. The positive
tectomy and breast implant placement. Silicone specific MR image predictive value of BI-RADS microcalcification descriptors
obtained in the sagittal plane demonstrates a radial fold (arrow) and final assessment categories. AJR Am J Roentgenol
without implant rupture. 2010;194:1378—83.
[9] Leung JWT, Sickles EA. Developing asymmetry identi-
fied on mammography: correlation with imaging outcome
Water droplets. The ‘‘water droplets’’ descriptor has and pathologic findings. AJR Am J Roentgenol 2007;188:
been created for water droplets associated with a breast 667—75.
implant. [10] Chang CB, Lvoff NM, Leung JW, Brenner RJ, Joe BN, Tso
Peri-implant fluid. ‘‘Peri-implant fluid’’ is a new descrip- HH, et al. Solitary dilated duct identified at mammogra-
tor available to describe fluid surrounding a breast implant. phy: outcomes analysis. AJR Am J Roentgenol 2010;194:
378—82.
[11] Destounis S, Arieno A, Morgan R, Murphy P, Seifert P,
Funding Somerville P, et al. Clinical experience with elasticity imag-
ing in a community-based breast center. J Ultrasound Med
This research did not receive any specific grant from funding 2013;32:297—302.
[12] Chang JM, Won J, Lee K, Park IA, Yi A, Moon WK. Comparison
agencies in the public, commercial, or not-for-profit sectors.
of shear-wave and strain ultrasound elastography in the dif-
ferentiation of benign and malignant breast lesions. AJR Am J
Roentgenol 2013;201:W347—56.
Disclosure of interest [13] Ozsoy A, Acar D, Barca AN, Aktas H, Araz L, Ozkaraoglu O,
et al. Diagnostic performance of real-time strain sonoelastog-
The authors declare that they have no competing interest. raphy in BI-RADS 4 and 5 breast masses. Diagn Interv Imaging
2016;97:883—9.
[14] Telegrafo M, Rella L, Stabile Ianora AA, Angelelli G, Moschetta
References M. Effect of background parenchymal enhancement on breast
cancer detection with magnetic resonance imaging. Diagn
[1] American College of Radiology. ACR BI-RADS atlas: breast imag- Interv Imaging 2016;97:315—20.
ing reporting and data system. 5th ed. Virginia: Reston; 2013.

Downloaded for Anonymous User (n/a) at University Of Texas Southwestern Medical Center At Dallas from ClinicalKey.com by Elsevier on August 28, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
View publication stats