FOR HKMA CME MEMBER ARTICLE
USE ONLY.IN
DOPRESS
NOT REPRODUCE OR DISTRIBUTE
The Breast (2006) 15, 313–318
THE BREAST
www.elsevier.com/locate/breast
REVIEW
Fat necrosis of the breast—A review
P.H. Tan, L.M. Lai, E.V. Carrington, A.S. Opaluwa, K.H. Ravikumar,
N. Chetty, V. Kaplan, C.J. Kelley, E.D. Babu
Department of Surgery, Hillingdon Hospital NHS Trust, Pield Health Road, Uxbridge,
Middlesex, UB8 3NN, UK
Received 22 April 2005; accepted 20 July 2005
KEYWORDS
Fat necrosis;
Breast;
Fine needle
aspiration cytology;
Core biopsy;
Radiology
Summary Fat necrosis of the breast is a benign condition that most frequently
affects peri-menopausal women. It can mimic breast cancer clinically or
radiologically. In other cases it can obscure malignant lesions.
The core of this review is derived from a MEDLINE database literature search from
1966–2004. Further references were from lateral search.
In this paper, we review the pathogenesis and pathology clinical and radiological
features of fat necrosis of the breast. The implication of fat necrosis in the
management of patients with breast lump is also discussed.
Fat necrosis of breast is a complex process. Therefore, a systematic review of this
condition will enable surgeons, radiologists and oncologists working in the field of
breast disease to understand it better and improve its management.
& 2005 Elsevier Ltd. All rights reserved.
Introduction
Fat necrosis is a benign non-suppurative inflammatory process of adipose tissue which was initially
described in the breast in 1920.1,2 Hadfield described it as ‘‘an innocent lesion of the breast
presenting itself most often in women between the
fourth and fifth decades, frequently as a stonyhard tumour firmly fixed to the skin, often
resembling an early cancer so closely that a wide
resection of the breast has been performed’’.3
Corresponding author. Tel.: +01895 279 156;
fax: +01895 279 156
E-mail address: [email protected] (E.D. Babu).
It is important to diagnose fat necrosis because it
often mimics carcinoma of the breast. The aim of
this paper is to review the clinical, pathological and
radiological features of fat necrosis of the breast
which distinguishes it from breast cancer.
Epidemiology
The incidence of the disease is estimated to
be 0.6% in the breast, representing 2.75% of
all benign lesions.1–4 Fat necrosis is found in 0.8%
of breast tumours and 1% of breast reduction surgery cases.5 The average age of patients is
50 years.1–4
0960-9776/$ - see front matter & 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.breast.2005.07.003
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE
ARTICLE
IN NOT
PRESS
FOR HKMA CME MEMBER USE
ONLY. DO
REPRODUCE OR DISTRIBUTE
314
P.H. Tan et al.
Aetiology
The
aetiological
factors
include
trauma
(21–70%),1–4 radiotherapy,6–8 anticoagulation (warfarin),9 cyst aspiration, biopsy, lumpectomy, reduction mammoplasty, implant removal, breast
reconstruction with tissue transfer,10 duct ectasia
and breast infection. Other rare causes include
polyarteritis nodosa, Weber-Christian disease and
granulomatous angiopanniculitis. In some patients,
the cause is unknown.11
Pathogenesis
Fat necrosis is a sterile, inflammatory process
which results from aseptic saponification of fat by
means of blood and tissue lipase.12 It varies in
appearance depending on the stage of the process.
It is recognised histologically as fat-filled macrophages and foreign body giant cells surrounded by
interstitial infiltration of plasma cells.13
After trauma to the breast, haemorrhage occurs
within a surgical cavity, or extravasate into the
parenchyma. Blood dissects along the fibrous
planes of breast (contusion). Contusion may also
occur secondary to capillary rupture, thrombophilia, allergic reaction, and arterial or venous
thromboses.
Extravasation of blood into the parenchyma
causes oedema and swelling of the trabecular
framework of the breast, with considerable tissue
ischaemia and pressure necrosis, with subsequent
disruption and fragmentation of fat cells.
This destruction causes formation of intra-cellular vacuoles filled with necrotic lipid material.
Fibroblasts, multinucleated giant cells and lipidladen histiocytes (‘fat-filled macrophages’ or ‘foam
cells’) accumulate between cyst-like areas.
At a cellular level, apoptosis and necrosis are two
forms of cell death with distinctive morphological
and biochemical features.14 Apoptosis is a process
of ‘programmed cell death’ while necrosis is a
deranged form of cell death. The intensity of the
initial insult determines whether apoptosis or
necrosis predominates. Various stimuli such as
cytokines, ischemia, heat, irradiation and pathogens can trigger both apoptosis and necrosis.14
Furthermore, signalling pathways such as death
receptors, kinase cascades, and mitochondria,
participate in both processes.14
In necrosis, cytosolic constituents spill into the
extracellular space through the cell membrane and
provoke an inflammatory response. In apoptosis
these products are safely isolated by cell mem-
branes and are then consumed by macrophages.15
Both processes prevail in fat necrosis of the breast.
A fine balance between apoptosis and necrosis
determines the clinical and radiological appearance in fat necrosis of the breast. A greater
necrotic component results in more inflammation
and hence a more sinister presentation of the
condition.
Healing occurs by fibrosis which begins at
the periphery of the cyst-like areas. Depending
upon the degree of fibrosis, these areas are
either replaced completely by fibrous tissue or
remains as a cavity.10,16 Calcifications occur in the
area of fibrosis and hence are a relatively late
finding.17
Clinical features
Clinical presentation of fat necrosis can range from
an incidental benign finding to a lump highly
suggestive of cancer.16,18 In most cases it is
clinically occult; however, it can present as single
or multiple smooth, round, firm nodules or irregular
masses. It may be associated with ecchymosis,
erythema, inflammation, pain, skin retraction or
thickening, nipple retraction and lymphadenopathy
simulating carcinoma.10,16,19,20
Hadfield3 described 66% of lesions as being
stony hard. Up to 52% of these lesions are adherent
to the skin and 10% are associated with retraction
of the nipple. Adair and Munzer2 described 64% of
patients with skin tethering. Pullyblank et al.4
found clinical signs suggestive of cancer in 45% of
cases.
There appears to be little difference in clinical
presentations of lesions regardless of whether they
are related to trauma.3,4 In cases related to
trauma, the majority of patients presented with a
lump in the breast. The mean time for patients to
present with a breast lump from the time of trauma
is 68.5 weeks (range 3–208).4
Both Hadfield3 and Adair and Munzer2 found that
in lesions precipitated by trauma, the lump
occurred at the site of the trauma. In cases where
there was no history of trauma, Hadfield3 found the
lesion to be most often in the upper outer quadrant
of the breast. The fact that most patients with
trauma had a lump in the upper inner quadrant is
thought to be related to the use of seatbelts. DiPiro
et al.21 described different patterns of injuries to
the inner breast according to whether the patient
was the passenger or the driver of the vehicle. Fat
necrosis is commonly found in the superficial breast
tissues and subareolar regions in obese women with
pendulous breasts.10
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE
FOR HKMA CME MEMBER ARTICLE
USE ONLY.IN
DOPRESS
NOT REPRODUCE OR DISTRIBUTE
Fat necrosis of the breast—A review
Tissue diagnosis
Fine needle aspiration cytology (FNAC)
FNAC is reported to have a high sensitivity and
specificity—87% and 99%, respectively.22,23
Using FNAC as a diagnostic tool enables rapid
results particularly in a ‘one-stop’ clinic setting.
FNAC can decrease the number of excision biopsies.22,23 However, the diagnosis of fat necrosis
using FNAC is limited by inadequate samples24 and
repeated attempts may be necessary to achieve a
confident result. FNAC is reliable in collaboration
with a good history of trauma and close follow-up in
diagnosing fat necrosis.25
Core biopsy
315
fibrous capsule. They have benign looking round or
oval, radiolucent shapes (sharply demarcated lucent area) with a thin rim on mammograms. Older
lesions appear as radiolucent oil cysts with ring-like
dystrophic calcifications in the wall. Fat necrosis
may also appear as lipophagic granulomas which
are associated with greater degree of fibrosis.
Occasionally fat necrosis may manifest as
ill-defined, spiculated dense or focal masses
associated with distortion, skin thickening and
retraction as commonly seen in breast cancer.10,16,19,21,29,30 Fat necrosis may also be associated with pleomorphic, clustered or branchingtype microcalcifications which are indistinguishable
from those associated with cancer.
Ultrasound
Core biopsy of breast lesions has been shown to be
more sensitive than FNAC.6 The diagnostic accuracy
between large-core needle biopsy and excision
(surgical) biopsy are favourably comparable.26,27
Parker et al.26 report false negative rates of
1.2–1.5% with large-core needle biopsy. However,
core biopsy is often inadequate or indeterminate in
the investigation of suspected fat necrosis in the
breast24,25 Pullyblank et al.4 reported two (4.7%)
cases in their series of 42 patients who were found
to have malignancy on excision biopsy after a core
biopsy diagnosis of fat necrosis. Hence surgical
excision biopsy is indicated where there is still
suspicion of malignancy following negative core
biopsy.
Imaging
Mammography
The mammographic appearance of fat necrosis
include normal appearance (9%), discrete round
or oval radiolucent oil cyst with thin capsule (27%),
thickening and deformity of skin and subcutaneous
tissue (16%), focal mass (13%), and ill-defined
spiculated mass (4%).5 Oil cysts may be associated
with uniform continuous eggshell calcification
(27%). There may also be multiple clustered
pleomorphic microcalcifications (4%) suspicious of
malignancy.10,19,28 The most common mammographic findings are dystrophic calcifications, followed by radiolucent oil cysts.
The variable mammography changes are attributable to the degree of fibrosis in different stages of
the disease. During early stages, less extensive
fibrosis is associated with a lipid cyst with a thin,
The ultrasound appearance of fat necrosis range
from solid nodules with posterior shadowing, to
complex intra-cystic masses that evolve over
time.18,30 These features depict the histological
evolution of fat necrosis.11,31 Sonographically, fat
necrosis may appear as cystic or solid masses.
Cystic lesions appear complex with mural nodules
or with internal echogenic bands. Solid masses have
well-circumscribed or ill-defined margins, and are
often associated with distortion of the breast
parenchyma.25
Common features of fat necrosis on ultrasonogram are increased echogenicity of subcutaneous
tissue (27%), as an anechoic cyst with posterior
acoustic enhancement (17%), hypoechoic mass with
posterior acoustic shadowing (16%), solid mass
(14%), cyst with internal echoes (11%), normal
appearance (11%) or cystic mural nodule
(4%).5,18,30,31
The oil cyst, which shows posterior acoustic
shadowing, corresponds to the round radiolucent
lesion with curvilinear wall calcification on mammography. Ultrasound can reliably diagnose oil
cysts. Soo et al.18 defined an echogenic band that
shifted in orientation with changes in patient
position as being diagnostic of oil cysts.
Magnetic resonance imaging (MRI)
Fat necrosis produces a wide spectrum of findings
on MRI. Magnetic resonance images correlate well
with the histology of fat necrosis. The abundant
iron-containing siderophages cause a diffuse decrease in signal intensity on both T1- and T2weighted images. A focal area of oedema is seen as
a hyperintense zone on T2-weighted images.
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE
ARTICLE
IN NOT
PRESS
FOR HKMA CME MEMBER USE
ONLY. DO
REPRODUCE OR DISTRIBUTE
316
P.H. Tan et al.
Fat necrosis consists of oil cysts and lipophagic
granulomas in varying proportions. Pure oil cysts
are fairly well identified as round, well-circumscribed hyperintense areas on T1-weighted images.
They may show a faint rim of enhancement along
the boundary of the cavity. Lipophagic granulomas
are more difficult to distinguish from malignancy on
MRI.32,33 Hence such lesions require biopsy for
confirmation of diagnosis.
MRI may show irregular peripheral enhancement
with a non-enhancing central areas in fat necrosis.34 Histologically the enhancing areas correspond
to peripherally developing fibrosis and inflammatory cell infiltration, and the central non-enhancing
areas correspond to necrotic fat. However, this MR
appearance of fat necrosis may be indistinguishable
from malignant lesions which also often have
necrotic centres.13,35–38 Fat necrosis is known to
mimic the appearance of recurrent breast cancer
on MRI.36–39 It can also appear as spiculated
enhancement without a washout surrounding a
cystic lesion with a fat-fluid level,38 or a cystic
lesion with a fat-fluid level and without enhancement13 Villeirs et al.40 reported a rare case of fat
necrosis after heparin-induced thrombocytopenia,
in which MRI demonstrated a diffuse early enhancement without a washout.41
Fresh granulation tissue is highly vascular and
may enhance significantly, therefore, it is generally accepted that MR is not useful during the
first 6 months after trauma or surgery. After
this period, hypovascular scar tissue develops,
with little or no enhancement, providing an
excellent distinction between old scarring and
malignancy.32,33
Discussion
The management of fat necrosis continues to be
challenging in practice. Even with modern diagnostic modalities, fat necrosis of the female breast
can still be difficult to diagnose. In patients who
have undergone breast conservation surgery or
reconstruction for breast cancer, fat necrosis must
be distinguished from cancer recurrence. Hence in
specific cases, needle core biopsy is required to
confirm diagnosis.
Although there is a definite association with
trauma, surgery or biopsy of the breast, not all
patients present with a clear history for fat
necrosis.42 As the commonest presentation is that
of a lump there, an underlying malignancy must be
considered.43 Even with a clear history of previous
trauma, the possibility of a malignancy should not
be overlooked, as patients’ attention may only be
drawn to the lump by an episode of trauma.44
Management may include short-term follow-up
with imaging and physical examination, to reduce
the number of unnecessary biopsies. In one series
on the mammographic follow-up period of 3 years,
it was noted that six round opacities decreased in
size and density, and in two of them, a radiolucent
oil cyst and calcifications developed. Eleven dystrophic calcifications changed to a benign appearance.4 In ultrasonographic follow-up, the most
common finding was normalisation of the subcutaneous tissue echogenicity and formation of small
cysts.18 Solid masses remained solid, whereas
complex lesions tended to evolve. Most of the solid
masses decreased in size confirming a benign
process. None of the complex lesions became more
solid, while some became more cystic. No mass
enlarged on ultrasound follow-up in this series.
Although the changes in echo texture on ultrasound may be chronologically and morphologically
consistent with the diagnosis of fat necrosis, one
must be wary of sinister pathology obscured by
increased density on mammography. Multiplicity
and distribution along the seat-belt line may
warrant follow-up rather than a biopsy, particularly
in young patients.30
The recent work of Soo et al.18 concluded that
with the superior quality of current imaging, any
abnormality with sufficiently benign characteristics
might safely be managed expectantly. A needle
core biopsy may be reserved for selected cases
where the lesion appears indeterminate or suspicious. Clearly, this strategy involves a period of
observation before the lesion can be confirmed as
benign which may well exacerbate patient anxiety.
Furthermore, 66% of mammographic abnormalities
and 74% of ultrasound lesions were classified as
indeterminate or suspicious. For best practice, the
histological result should be considered in conjunction with the clinical and imaging findings, with
involvement of a multidisciplinary team.
In conclusion, there is a wide range of manifestations of fat necrosis clinically and radiologically.
Knowledge of the appearance and evolution of
these patterns and a careful investigation of the
history would enable prompt diagnosis and appropriate management i.e. follow-up or excision
biopsy. It has to be noted that patients who present
to breast symptomatic clinics with carcinoma may
give a history of trauma. It has been noted that
the manifestations of breast trauma are often
suspicious of carcinoma on clinical and imaging
grounds. This reinforces the need for initial triple
assessment. In view of the complexity of clinical
and radiological findings at presentation, the
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE
FOR HKMA CME MEMBER ARTICLE
USE ONLY.IN
DOPRESS
NOT REPRODUCE OR DISTRIBUTE
Fat necrosis of the breast—A review
investigation and management of fat necrosis does
not fit easily into a rigid protocol. Where there is
any diagnostic uncertainty we recommend needle
core biopsy to obtain definitive diagnosis and that
follow-up imaging is performed to confirm resolution. Pitfalls in diagnosis of fat necrosis may arise in
cases following radiotherapy and/or surgery for
breast carcinoma. Therefore the diagnosis of fat
necrosis should only be a diagnosis by exclusion of
malignancy.
317
20.
21.
22.
23.
References
24.
1. Lee BJ, Adair F. Traumatic fat necrosis of the female breast.
Ann Surg 1920;72:188–95.
2. Adair FE, Munzer JT. Fat Necrosis of the female breast:
report of one hundred test cases. Am J Surg 1947;74:
117–28.
3. Hadfield G. Fat necrosis of the breast. Br J Surg 1930;17:
673–82.
4. Pullyblank AM, Davies JD, Basten J, Rayter Z. Fat necrosis of
the female breast—Hadfield re-visited. Breast 2001;10(5):
388–91.
5. Pui MH, Movson IJ. Fatty tissue breast lesions. Clin Imag
2003;27(3):150–5.
6. Stefanik DF, Brereton HD, Lee TC, Chun BK, Citgay O,
Dritschilo A. Fat necrosis following breast irradiation for
carcinoma; clinical presentation and diagnosis. Breast
1982;8:4–6.
7. Clarke D, Curtis JL, Martinez A, Fajardo L, Goffinet D. Fat
necrosis of the breast simulating recurrent carcinoma after
primary radiotherapy in the management of early stage
breast carcinoma. Cancer 1983;52(3):442–5.
8. Rostom AY, El-Sayed ME. Fat necrosis of the breast: an
unusual complication of lumpectomy and radiotherapy in
breast cancer. Clin Radiol 1987;38:31.
9. Lopez Valle CA, Hebert G. Warfarin-induced complete
bilateral breast necrosis. Br J Plast Surg 1992;45:606–9.
10. Hogge JP, Robinson RE, Magnant CM, Zuurbier RA. The
mammographic spectrum of fat necrosis of the breast.
Radiographics 1995;15(6):1347–56.
11. Chala LF, de Barros N, de Carmargo Moraes P, Endo E,
Pincetrato KM, Carvalho FM, et al. Fat necrosis of the
breast: mammographic, sonographic, computed tomography, and magnetic resonance imaging findings. Curr Probl
Diagn Radiol 2004;33:106–26.
12. Frates MC, Homer MJ, Robert NJ, Smith TJ. Noniatrogenic
breast trauma. Breast Dis 1992;5:11–9.
13. Youssefzadeh S, Wolf G, Imhof H. MR findings of a breast oil
cyst containing a fat-fluid level. A case report. Acta Radiol
1994;35(5):492–4.
14. Ellis RE, Yuan JY, Horvitz HR. Mechanisms and functions of
cell death. Annu Rev Cell Biol 1991;7:663–98.
15. Srivastava P. Roles of heat-shock proteins in innate and
adaptive immunity. Nat Rev Immunol 2002;2(3):185–94.
16. Evers K, Troupin RH. Lipid cyst: classic and atypical
appearances. AJR Am J Roentgenol 1991;157(2):271–3.
17. Cyrlak D, Caepenter PM. Breast imaging case of the day.
Radiographics 1999;19:S80–3.
18. Soo MS, Kornguth PJ, Hertzberg BS. Fat necrosis in the
breast: sonographic features. Radiology 1998;206(1):261–9.
19. Bassett LW, Gold RH, Cove HC. Mammographic spectrum of
traumatic fat necrosis: the fallibility of ‘‘pathognomonic’’
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
signs of carcinoma. AJR Am J Roentgenol 1978;130(1):
119–22.
Orson LW, Cigtay OS. Fat necrosis of the breast: characteristic xeromammographic appearance. Radiology 1983;
146(1):35–8.
DiPiro PJ, Meyer JE, Frenna TH, Denison CM. Seat belt
injuries of the breast: findings on mammography and
sonography. AJR Am J Roentgenol 1995;164(2):317–20.
De Freitas Jr R, Hamed H, Fentiman I. Fine needle aspiration
cytology of palpable breast lesions. Br J Clin Pract 1992;
46(3):187–90.
Brown LA, Coghill SB. Fine needle aspiration cytology of the
breast: factors affecting sensitivity. Cytopathology 1991;
2(2):67–74.
Dinakara Babu E, Balasubramaniam B, Dawson JW. Triple
assessment of fat necrosis of the breast. Eur J Surg Oncol
2001;27(8):816.
Haagensen CD. Traumatic fat necrosis in the breast, 2nd ed.
Philadelphia: Saunders; 1971.
Parker SH, Burbank F, Jackman RJ, Aucreman CJ, Cardenosa
G, Cink TM, et al. Percutaneous large-core breast biopsy: a
multi-institutional study. Radiology 1994;193(2):359–64.
Parker SH, Jobe WE, Dennis MA, Stavros AT, Johnson KK,
Yakes WF, et al. US-guided automated large-core breast
biopsy. Radiology 1993;187(2):507–11.
Bilgen IG, Ustun EE, Memis A. Fat necrosis of the breast:
clinical, mammographic and sonographic features. Eur J
Radiol 2001;39(2):92–9.
Leborgne R. Calcified cystic steatonecrosis of the breast.
Radiological study. Torax 1967;16(3):172–5.
Sullivian TP, Georgian-Smith D. Breast mass detected
after blunt chest trauma. AJR Am J Roentgenol 1998;
171:50.
Harvey JA, Moran RE, Maurer EJ, DeAngelis GA. Sonographic
features of mammary oil cysts. J Ultrasound Med
1997;16(11):719–24.
Heywang-kobrunner SH, Beck R. Contrast-enhanced MRI
of breast, 2nd ed. Berlin Heidelberg, New York: Springer;
1995.
Friedrich M. Magnetic resonance imaging of the breast. In:
Friedrich M, Sickles EA, editors. Radiological diagnosis of
breast disease. Berlin Heideberg New York: Springer; 1997.
p. 243–54.
Kinoshita T, Yashiro N, Yoshigi J, Ihara N, Narita M. Fat
necrosis of breast: a potential pitfall in breast MRI. Clin Imag
2002;26(4):250–3.
Pierce WB, Harms SE, Flamig DP, Griffey RH, Evans WP,
Hagans JE. Three-dimensional gadolinium-enhanced MR
imaging of the breast: pulse sequence with fat suppression
and magnetization transfer contrast. Work in progress.
Radiology 1991;181(3):757–63.
Gilles R, Guinebretiere JM, Shapeero LG, Lesnik A, Contesso
G, Sarrazin D, et al. Assessment of breast cancer recurrence
with contrast-enhanced subtraction MR imaging: preliminary
results in 26 patients. Radiology 1993;188(2):473–8.
Cohen EK, Leonhardt CM, Shumak RS, Soutar IC, Bukhanov K,
Fishell EK, et al. Magnetic resonance imaging in potential
postsurgical recurrence of breast cancer: pitfalls and
limitations. Can Assoc Radiol J 1996;47(3):171–6.
Coady AM, Mussurakis S, Owen AW, Turnbull LW. Case report:
MR imaging of fat necrosis of the breast associated with lipid
cyst formation following conservative treatment for breast
carcinoma. Clin Radiol 1996;51(11):815–7.
Solomon B, Orel S, Reynolds C, Schnall M. Delayed
development of enhancement in fat necrosis after
breast conservation therapy: a potential pitfall of MR
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE
ARTICLE
IN NOT
PRESS
FOR HKMA CME MEMBER USE
ONLY. DO
REPRODUCE OR DISTRIBUTE
318
P.H. Tan et al.
imaging of the breast. AJR Am J Roentgenol 1998;
170(4):966–8.
40. Villeirs G, Van Damme S, Heydanus R, Serreyn R, Kunnen M,
Mortier M. Heparin-induced thrombocytopenia and fat
necrosis of the breast. Eur Radiol 2000;10(3):527–30.
41. Whitehouse GH, Moore NR. MR imaging of the breast after
surgery for breast cancer. Magn Reson Imag Clin N Am
1994;2(4):591–603.
42. Tohno E, Cosgrove DO, Sloane JP. Benign processes: trauma
and iatrogenic conditions. In: Ultrasound diagnosis of
breast. Stutgard: Thieme; 1994. p. 338.
43. Farrow JH. Common benign lesions of the adult human
breast. Clin Obstet Gynaecol 1966;9:170–85.
44. Minagi H, Youker JE. Roentgen appearance of fat necrosis in
the breast. Radiology 1968;90(1):62–5.
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE