BI-RADS update for breast cancer caregivers

REVIEW

BI-RADS update for breast cancer caregivers

Emily L. Sedgwick1• Lilian Ebuoma1

• Anthony Hamame1• Kanchan Phalak1

•

Lorell Ruiz-Flores1• Tamara Ortiz-Perez1

• Karla A. Sepulveda1

Received: 2 March 2015 / Accepted: 4 March 2015 / Published online: 12 March 2015

� Springer Science+Business Media New York 2015

Abstract This review will discuss changes relevant to

breast cancer caregivers in the fifth edition of the Breast

Imaging Reporting and Data System.

Keywords BI-RADS � Breast imaging � Mammography �MRI � Ultrasound

The fifth edition of the Breast Imaging Reporting and Data

System (BI-RADS) impacts not only the breast imaging ra-

diologist and the referring breast care provider, but also the

patient. It is incumbent upon all health care workers involved

in breast care to understand the changes of the new edition of

BI-RADS because it affects how we communicate with one

another, how we make management decisions, and how we

use this information in research and quality audits to provide

future care.

The BI-RADS lexicon, a product of the American College

of Radiology, was mandated by the Mammographic Quality

and Standard Act of 1992. The Act pertains solely to mam-

mography. The mammography lexicon is a dynamic entity,

now in its fifth iteration. Because of the success of the

mammography lexicon, the breast ultrasound and breast

MRI components were subsequently created. The fifth edi-

tion of the BI-RADS mammography lexicon is coupled with

the 2nd editions of the breast ultrasound and breast MRI

lexicons. In this version, standardization of the lexicon

across the different breast imaging modalities was performed

to simplify the lexicon for easier integration into everyday

use.

The changes in the new edition can be categorized into 3

groups: lexicon, reporting, and management. Changes will

be reviewed, and images will be provided to further illus-

trate the concept. Where available, data will be provided to

support the change.

Mammography

Lexicon

The lexicon was originally created by a group of breast

imaging experts. Although much of the lexicon continues to

rely upon expert opinion, the lexicon is increasingly founded

upon evidence. Evidence to support lexicon changes is most

robust in the mammography section, largely due to its early

creation. The first group of changes was to the words used to

describe mammographic findings (‘‘descriptors’’). Mam-

mographic findings are grouped into seven categories:

masses, calcifications, architectural distortion, asymmetries,

intramammary lymph node, skin lesions, and solitary dilated

duct. Asymmetries, intramammary lymph node, skin lesions,

and solitary dilated duct are newly defined categories. See

Figs. 1, 2, 3, and 4. Under masses, the term ‘‘lobulated’’ has

been eliminated. In the category of calcifications, three

changes were made: the terms ‘‘lucent-centered’’ and ‘‘egg

shell’’ were eliminated, and the term ‘‘group’’ exchanged for

‘‘cluster’’ under the subcategory of calcification distribution.

A group of calcifications is defined as 5 calcifications within

1 cm area of tissue, but less than 2 cm [1]. A collection of

calcifications larger than 2 cm in a non-segmental distribu-

tion is called ‘‘regional calcifications.’’ See Figs. 5 and 6.

& Emily L. Sedgwick

[email protected]

1 Department of Radiology, Baylor College of Medicine,

1 Baylor Plaza, MS-360, Houston, TX 77030, USA

123

Breast Cancer Res Treat (2015) 150:243–254

DOI 10.1007/s10549-015-3332-4

Reporting

A change in the BI-RADS reporting that will be noticeable

for referring clinicians is the elimination of numeric values

assigned to breast tissue composition. In an effort to dis-

cern breast density from report assessments (both of which

used ‘‘BI-RADS 1 through 4’’), breast density will now be

described as BI-RADS A (fatty) through D (extremely

dense). Due to significant overlap in radiologists’ assess-

ment of adjacent categories, the former numeric quantifi-

cation of dense breast tissue (e.g.,\25 % dense tissue) has

been eliminated [1].

Proper characterization of the morphology and distri-

bution of calcifications is fundamental to the usefulness of

the lexicon. The PPV (likelihood of malignancy) has been

studied which has resulted in changes in this edition of BI-

RADS. It has been shown that the PPV of abnormal cal-

cifications on screening mammography of women who

were ultimately diagnosed with breast cancer was 19.1 %

[2]. Specific descriptors have been studied by multiple

investigators, four of which have relevance to BI-RADS:

coarse heterogeneous, amorphous, fine pleomorphic, and

fine linear branching. Coarse heterogeneous calcifications

had PPVs ranging from 7 to 50 % [3, 4]. Amorphous cal-

cifications had PPVs that ranged from 13 to 26 % [3–6].

The PPVs of fine pleomorphic calcifications have been

reported to be approximately 28 % [4]. Fine linear

branching calcifications have the highest PPVs, ranging

from 53 to 81 % [3, 4, 6]. The reporting of BI-RADS 4

(‘‘Findings are suspicious for malignancy, biopsy should be

considered’’) had been previously subdivided into 4a-c to

assist the referring physician in the decision to perform a

biopsy. This edition assigns PPVs to the subdivisions: 4a

(PPV of[2–10 %), 4b ([10–50 %), 4c ([50 % to\95 %).

Although the assignment of the 4a-c subdivisions is still

voluntary, greater adoption is expected. For example, fine

Fig. 1 a, b Note the asymmetry

in the central right breast

(arrows) as compared to the

same location in the left central

breast

244 Breast Cancer Res Treat (2015) 150:243–254

123

linear branching calcifications should be given a 4c as-

signment, as the PPVs range from 53 to 81 %. See Fig. 7.

Two other changes are relevant to the referring physi-

cian. The first is to include the reporting of a specific lo-

cation of a finding which includes laterality, quadrant,

clock face orientation, depth, and distance from the nipple

(e.g., ‘‘left upper outer quadrant at 2 o’clock, middle depth,

5 cm from the nipple’’). Specific location descriptions can

be especially helpful in the absence of prior images. Se-

cond, a screening mammogram should not be assessed as

BI-RADS 3, 4 or 5. Those assessments are reserved for

diagnostic modalities only (diagnostic mammography, ul-

trasound or breast MRI). This recommendation is now one

of the Physician Quality Reporting System measures from

the Centers of Medicare and Medicaid Services.

Management

A significant, and controversial, change relevant to all

breast care providers is the separation of the BI-RADS

assessment from management recommendations. Formerly,

the BI-RADS assessment directed management (e.g., BI-

RADS 3 usually indicated a 6-month follow up). Radi-

ologists, however, have been increasingly faced with the

dilemma of a BI-RADS 3 assessment but then performing a

biopsy at the request of the patient or referring provider.

Since the BI-RADS audit has driven improvement, the

Fig. 2 a Intramammary lymph

node (box) in the upper outer

breast, a common location for

an intramammary lymph node.

b Intramammary lymph node

(arrow). Note the fat in the

center of the lymph node, a

characteristic feature

Fig. 3 Dilated duct

Breast Cancer Res Treat (2015) 150:243–254 245

123

integrity of the audit can be preserved by separating the

assessment from the management. Many existing software

programs use the assessment categories to perform both the

Fig. 4 a, b Symmetric bilateral

skin thickening and

parenchymal edema in a patient

with congestive heart failure

Fig. 5 Grouped calcifications are a collection of calcifications

spanning 2 cm or less

Fig. 6 Regional calcifications are a collection of calcifications that

span more than 2 cm of breast tissue


123

audit and drive appropriate follow up. Future reporting

software will have to create algorithms which permit logs

for appropriate patient follow up and to perform audits

where assessment and management are independent

variables.

One controversy involves the use of BI-RADS 0 and

problem solving breast MRI. Although the use of breast

MRI as a problem solving tool is infrequent, breast MRI

can be a valuable tool in patients who have had breast

conservation therapy and who perceive changes in the

lumpectomy bed. Diagnostic mammography and ultra-

sound are of little use due to the expected post-operative

changes, and breast MRI has been found to be useful to

differentiate scar from tumor [7]. The new BI-RADS

Fig. 7 Fine, linear branching calcifications should be given a BI-

RADS 4c assessment

Fig. 8 Amorphous calcifications should be given a BI-RADS 4b

assessment

Fig. 9 a–c Developing asymmetry (arrow). Note the asymmetry development over time

Fig. 10 Intraductal calcifications (arrows)


123

edition strongly discourages the use of BI-RADS 0 in this

circumstance; however, many radiologists are reluctant to

assign a BI-RADS assessment to the mammogram without

the additional information provided by the breast MRI.

Since the use of the BI-RADS lexicon is voluntary, there

may be practitioner variation in this patient scenario. It is

important to communicate with your local radiologist about

how this situation is managed at your facility.

Two biopsy recommendations relevant to referring

physicians are the recommendations for biopsy of amor-

phous calcifications and for a developing asymmetry. As

mentioned above, the PPV of a biopsy recommendation of

amorphous calcifications ranges between 13 and 26 % [3–

6]. Formerly, amorphous calcifications were considered of

intermediate suspicion and may have been observed with

serial mammograms. Now, they would be assessed as BI-

RADS 4b, and biopsy would be recommended. See Fig. 8.

A developing asymmetry is a focal asymmetry which is new

or has suspicious characteristics (e.g., getting bigger). The

PPV is 12.8 %, so the BI-RADS assessment of a developing

asymmetry should be BI-RADS 4b [8]. See Fig. 9.

Ultrasound

Lexicon

Ultrasound findings are grouped into three categories:

masses, calcifications, and special cases. In the masses

Fig. 11 Simple cyst (arrow). Note the posterior feature of enhance-

ment (star)

Fig. 12 Vascular abnormality (pseudoaneurysm). Note the alternat-

ing directions of blood flow within the mass

Fig. 13 Mondor disease. Note the absence of blood flow within the

superficial thrombosed vein (arrow)


123

category, lesion boundary was eliminated. The posterior

acoustic features descriptor was changed to ‘‘posterior

features.’’ In the calcifications category, the macrocalcifi-

cations descriptor was eliminated, and intraductal (e.g.,

ductal carcinoma in situ) calcifications descriptor was

added. See Fig. 10. Several new Special Cases were added:

simple cyst, vascular abnormalities, Mondor disease,

postsurgical fluid collection, and fat necrosis. See Figs. 11,

12, 13, 14 and 15.

In this edition, a new sonographic technique was incor-

porated into BI-RADS: elasticity (as a subset of associated

features). Elasticity measures the hardness of a sonographic

mass via strain or shear wave. The more stiff the mass, the

more likelihood it is a cancer [9]. There is, however, sig-

nificant overlap between what is benign and what is ma-

lignant, so the tool is used to improve the specificity of a

biopsy recommendation [9, 10]. See Fig. 16.

Reporting

Currently, 19 states have adopted legislation regarding

breast density notification to women undergoing mam-

mography. Evidence supporting the use of screening

ultrasound in normal risk women is scant. In Connecti-

cut, where insurance coverage for additional screening

Fig. 14 a, b Postoperative fluid collections

Fig. 15 a, b Fat necrosis (arrows)


123

was mandated by law, the PPV for a finding seen only

on ultrasound was 6.5 % [11]. Due to the increased use

of ultrasound as a screening tool, this edition of BI-

RADS recommended making a tissue composition

assessment: homogeneous background echotexture (fat),

homogeneous background echotexture (fibroglandular),

and heterogeneous background echotexture. It is un-

known how the different echotextures affect the sensi-

tivity or specificity of screening ultrasound, a topic

likely to be addressed in the next edition of BI-RADS.

Fig. 16 a Elasticity-soft. Soft

masses are associated with

benignity (red circle).

b Elasticity-intermediate. Can

be malignant or benign.

c Elasticity-hard. Hard masses

are associated with malignancy

(red circle)


123

Management

Referring physicians should be aware of changes in the

recommended management of probable fibroadenomas and

complicated cysts. In prior editions, core needle biopsy or

aspiration had been the standard recommendation. For

probable fibroadenomas, ample evidence supports a BI-

RADS 3 (probably benign) assessment with 6-, 12-, and

24-month follow up ultrasounds [12, 13].

Breast magnetic resonance imaging (Breast MRI)

Lexicon

Breast MRI findings are grouped into seven categories for

dynamic contrast enhanced breast MRI (DCE-BMRI): fo-

cus, mass, nonmass enhancement, intramammary lymph

node, skin lesion, non-enhancing findings, and fat-con-

taining lesions. Associated features contain pertinent de-

scriptors for the 7 categories. Multiple descriptors have

been eliminated: foci, lobular, enhancing internal septation,

central enhancement, ductal enhancement, stippled, punc-

tate, reticular, dendritic, pre-contrast high duct signal,

edema, lymphadenopathy, hematoma-blood, and abnormal

signal void. Many of these terms were eliminated due to

underuse; others were eliminated in order to have homo-

geneity across the 3 lexicons. For example, mass margins

were now characterized as circumscribed (a benign de-

scriptor) versus non-circumscribed (a suspicious descrip-

tor) which is consistent with the mammography and

ultrasound lexicons. New descriptors include clustered ring

internal enhancement, intramammary lymph node and skin

lesions. See Figs. 17, 18 and 19. Finally, nonmass-like

enhancement category was simplified to nonmass

enhancement.

Fig. 17 Clustered ring enhancement (arrow). Sagittal T1 fat satura-

tion dynamic contrast enhanced subtraction image

Fig. 18 Intramammary lymph node (arrow). Note white fat in center

of the mass, a characteristic of an intramammary lymph node. Axial

T1 image

Fig. 19 Benign skin lesion-sebaceous cyst (arrow). Axial STIR

image


123

For the first time, silicone breast implants have been in-

cluded as a category in the BI-RADS lexicon. (Saline breast

implant integrity is a clinical assessment.) Silicone breast

implant integrity is assessed without contrast injection. De-

scriptors include type, location, abnormal implant contour,

intracapsular silicone implant findings, extracapsular silicone,

water droplets, and peri-implant fluid. See Figs. 20, 21, 22 and

23.

Reporting

Breast care health workers will now notice two standard phrases

at the beginning of the DCE-BMRI report regarding breast

tissue composition and background parenchymal enhance-

ment. Breast tissue composition is divided in the 4 categories, as

in mammography. Background parenchymal enhancement

(BPE) pattern is also divided into 4 categories (minimal, mild,

moderate, marked) and may be symmetric or asymmetric. See

Fig. 24. Inclusion of the BPE is important because there is

evidence that the degree of BPE may be associated with breast

cancer risk, like that of breast density on mammography [14].

Consistent use of the lexicon in the assessment of BPE will

allow for further investigation of this observation.

Kinetic assessment reporting has been further defined to

emphasize the importance of reporting both the initial and

delayed phases of the signal intensity changes over time.

Studies have increasingly shown the prediction of malig-

nancy in assessment of the initial phase, and some have

advocated fewer delayed contrast sequences to shorten the

imaging time [15].

Management

The most significant new management recommendation is

that any MRI finding that necessitates a second-look

ultrasound for biopsy localization should be given a BI-

RADS 4 or 5 assessment. In prior years, radiologists would

give a BI-RADS 0 assessment to indicate that additional

imaging was recommended. The BI-RADS 0 recommen-

dation, however, resulted in patients lost to follow up and

inaccurate quality auditing.

In summary, multiple changes have been made in the

fifth edition of BI-RADS which impacts our communi-

cation, patient management, research, and quality im-

provement. The lexicon has been simplified and

standardized across 3 breast imaging modalities to

Fig. 20 a Prepectoral silicone implant. Note the anterior location of

the implant (star) relative to the pectoralis major muscle (arrow).

Sagittal water saturation image. b Retropectoral silicone implant.

Note the posterior location of the implant (star) relative to the

pectoralis minor muscle (arrow). Sagittal T1 fat saturation dynamic

contrast enhanced image

Fig. 21 Focal implant bulge (arrow). Axial water saturation image of

a silicone breast implant


123

promote consistent use of the lexicon. Current reporting

software programs will need to evolve to accommodate

the separation of the BI-RADS assessment and the pa-

tient management recommendations to ensure adequate

patient follow-up. Frequent communication between

breast care providers and breast imaging radiologists is

encouraged for better understanding of the lexicon which

will foster improved patient care.

Fig. 22 a Intracapsular findings-subcapsular line (arrow). Sagittal

water saturation sequence. b Intracapsular findings-radial fold

(arrow). Axial STIR sequence. c Intracapsular findings-keyhole

(arrow). Axial water saturation sequence. d Intracapsular findings-

linguine sign (arrow). Sagittal water saturation sequence

Fig. 23 a Extracapsular findings-breast (arrow). Sagittal water saturation sequence. b Water droplets (arrow). Axial water saturation sequence


123

Conflict of interest The authors have no financial conflict of

interest.

Compliance with Ethical Standards The authors declare that this

review complies with United States’ ethical standards. As this is a

review, no human/animal experiments were performed, and no in-

formed consent was needed.

References

1. D’Orsi CJSE, Mendelson EB, Morris EA et al (2013) ACR BI-

RADS� Atlas, Breast Imaging Reporting and Data System.

American College of Radiology, Reston

2. Venkatesan A, Chu P, Kerlikowske K, Sickles EA, Smith-Bind-

man R (2009) Positive predictive value of specific mammo-

graphic findings according to reader and patient variables.

Radiology 250:648–657

3. Burnside ES, Ochsner JE, Fowler KJ et al (2007) Use of mi-

crocalcification descriptors in BI-RADS 4th edition to stratify

risk of malignancy. Radiology 242:388–395

4. Bent CK, Bassett LW, D’Orsi CJ, Sayre JW (2010) The positive

predictive value of BI-RADS microcalcification descriptors and

final assessment categories. Am J Roentgenol 194:1378–1383

5. Berg WA, Arnoldus CL, Teferra E, Bhargavan M (2001) Biopsy

of amorphous breast calcifications: pathologic outcome and yield

at stereotactic biopsy. Radiology 221:495–503

6. Liberman L, Abramson AF, Squires FB, Glassman JR, Morris

EA, Dershaw DD (1998) The breast imaging reporting and data

system: positive predictive value of mammographic features and

final assessment categories. Am J Roentgenol 171:35–40

7. Lee CH (2004) Problem solving MR imaging of the breast. Ra-

diol Clin North Am 42:919–934

8. Leung JW, Sickles EA (2007) Developing asymmetry identified

on mammography: correlation with imaging outcome and

pathologic findings. Am J Roentgenol 188:667–675

9. Mitka M (2007) New ultrasound ‘‘elasticity’’ technique may re-

duce need for breast biopsies. JAMA 297:455

10. Berg WA, Cosgrove DO, Dore CJ et al (2012) Shear-wave

elastography improves the specificity of breast US: the BE1

multinational study of 939 masses. Radiology 262:435–449

11. Hooley RJ, Greenberg KL, Stackhouse RM, Geisel JL, Butler RS,

Philpotts LE (2012) Screening US in patients with mammo-

graphically dense breasts: initial experience with Connecticut

Public Act 09-41. Radiology 265:59–69

12. Harvey JA, Nicholson BT, Lorusso AP, Cohen MA, Bovbjerg VE

(2009) Short-term follow-up of palpable breast lesions with be-

nign imaging features: evaluation of 375 lesions in 320 women.

Am J Roentgenol 193:1723–1730

13. Graf O, Helbich TH, Hopf G, Graf C, Sickles EA (2007) Prob-

ably benign breast masses at US: Is follow-up an acceptable al-

ternative to biopsy? Radiology 244:87–93

14. King V, Brooks JD, Bernstein JL, Reiner AS, Pike MC, Morris

EA (2011) Background parenchymal enhancement at breast MR

imaging and breast cancer risk. Radiology 260:50–60

15. Kuhl CK, Schrading S, Strobel K, Schild HH, Hilgers RD,

Bieling HB (2014) Abbreviated breast magnetic resonance

imaging (MRI): first postcontrast subtracted images and max-

imum-intensity projection-a novel approach to breast cancer

screening with MRI. J Clin Oncol 32:2304–2310

Fig. 24 a Background parenchymal enhancement-minimal. Axial fat

saturation dynamic contrast enhanced subtraction image. b Back-

ground parenchymal enhancement-mild. Axial fat saturation dynamic

contrast enhanced subtraction image. c Background parenchymal

enhancement-moderate. Axial fat saturation dynamic contrast en-

hanced subtraction image. d Background parenchymal enhancement-

marked. Axial fat saturation dynamic contrast enhanced subtraction

image


123

Documents

BI-RADS update for breast cancer caregivers