This copy is for personal use only. To order printed copies, contact reprints@rsna.
org
330
BREAST IMAGING
Idiopathic Granulomatous Mastitis:
Manifestations at Multimodality
Imaging and Pitfalls1
Cedric W. Pluguez-Turull, MD
Jennifer E. Nanyes, MD
Cristina J. Quintero, MD
Hamza Alizai, MD
Daniel D. Mais, MD
Kenneth A. Kist, MD
Nella C. Dornbluth, MD
Abbreviations: BI-RADS = Breast Imaging
Reporting and Data System, CC = craniocaudal,
FNA = fine-needle aspiration, IBC = inflamma-
tory breast cancer, IGM = idiopathic granulo-
matous mastitis, MLO = mediolateral oblique,
NME = nonmass enhancement, PAAG = poly-
acrylamide hydrogel
RadioGraphics 2018; 38:330–356
https://doi.org/10.1148/rg.2018170095
Content Codes:
1
From the Departments of Radiology (C.W.P.T.,
J.E.N., H.A., K.A.K., N.C.D.) and Pathology
(D.D.M.), University of Texas Health at San
Antonio, 7703 Floyd Curl Dr, San Antonio,
TX 78229; and Department of Radiology, Bryn
Mawr Hospital, Bryn Mawr, Pa (C.J.Q.). Pre-
sented as an education exhibit at the 2016 RSNA
Annual Meeting. Received April 13, 2017; revi-
sion requested August 10 and received Decem-
ber 21; accepted December 21. For this journal-
based SA-CME activity, the authors, editor, and
reviewers have disclosed no relevant relation-
ships. Address correspondence to C.W.P.T.
(e-mail: cedricpluguez@gmail.com).
©
RSNA, 2018
SA-CME LEARNING OBJECTIVES
After completing this journal-based SA-CME
activity, participants will be able to:
■■Describe the classic demographic fea-
tures, clinical manifestations, major dif-
ferential diagnoses, and histopathologic
findings of IGM.
■■Identifythe most common mammo-
graphic, US, and MR imaging findings
of IGM.
■■Compare the current options for non-
surgical versus surgical management of
IGM.
See www.rsna.org/education/search/RG.
Idiopathic granulomatous mastitis (IGM) is a rare benign inflam-
matory breast entity characterized by lobulocentric granulomas.
IGM has a persistent or recurrent disease course and affects parous
premenopausal women with a history of lactation. It has also been
associated with hyperprolactinemia. The most common clinical
sign is a palpable tender mass. However, the nonspecific manifesta-
tions and varied demographic features of this condition, as well as
the other similar-appearing and superimposed breast entities, pose
substantial diagnostic challenges. Entities with similar manifesta-
tions include inflammatory breast cancer (IBC), infective mastitis,
foreign body injection granulomas, mammary duct ectasia, diabetic
fibrous mastopathy, and systemic granulomatous processes. The
strategy for imaging IGM depends on patient age, clinical manifes-
tations, and risk factors. Targeted ultrasonography, mammography,
and less commonly, magnetic resonance imaging have proven to be
useful for imaging evaluation. Core-needle biopsy, with or without
fine-needle aspiration for cytopathologic examination, and culture
analysis are usually required to exclude IBC and other benign in-
flammatory breast processes. Patients with IGM have an excellent
prognosis when they are appropriately treated with oral steroids or
second-line immunosuppressive and prolactin-lowering medica-
tions. However, surgical excision may be an option for patients in
whom medication therapy is unsuccessful. Imaging surveillance can
be offered to patients with incidentally encountered IGM or mild
symptoms. Clinical suspicion for this rare disease and the breast
imager’s prompt diagnosis can lead to an improved patient outcome.
The purpose of this article is to review the imaging manifestations of
IGM in a multimodality case-based format and to describe relevant
clinical and imaging-based differential diagnoses. The associated pit-
falls, epidemiologic and histopathologic factors, clinical manifesta-
tions, natural course, and management of IGM also are discussed.
©
RSNA, 2018 • radiographics.rsna.org
Introduction
Idiopathic granulomatous mastitis (IGM), also known as nonpu-
erperal mastitis or granulomatous lobular mastitis, is a rare benign
chronic inflammatory breast disease that was first described by Kes-
sler and Wolloch (1) in 1972. IGM is characterized by sterile nonca-
seating lobulocentric granulomatous inflammation (1,2). It usually
has a recurrent or prolonged natural disease course that eventually
leads to lesion burnout (3–5). IGM usually affects parous premeno-
pausal women with a history of lactation and frequently is clinically
associated with hyperprolactinemia (6–9). Limited data regarding
the prevalence and incidence of IGM on the basis of ethnicity are
available; however, anecdotal experience and the few cases reported
in the literature suggest that the disease is rare (10). IGM can have a
wide range of clinical and imaging manifestations (10); however, the
RG  •  Volume 38  Number 2 Pluguez-Turull et al  331
TEACHING POINTS
■■ The results of several studies indicate that the most common
clinical manifestation of IGM is a tender palpable unilateral
breast mass of variable size (1–20 cm).
■■ The roles of the breast imager in the surveillance, presurgi-
cal, and posttreatment evaluation of confirmed IGM are to
(a) establish the multiplicity and location of IGM lesions,
(b) document the size of the lesion(s), (c) identify abscess for-
mation and the related possibility for intervention, (d) evaluate
the stability of or interval change in the lesion(s), (e) evaluate
the treatment response, and (f) identify metachronous disease
and local recurrence at imaging surveillance.
■■ Both a tissue specimen–based diagnosis of malignancy and
clinical evidence of inflammatory disease are required to con-
firm the diagnosis of IBC.
■■ Close surveillance alone may be appropriate for a subgroup
of patients, especially those in whom IGM is discovered inci-
dentally during screening mammography and those in whom
IGM manifests as a painless or mildly tender palpable mass.
■■ Corticosteroids have been shown to be an effective first-line
therapy for patients with histopathologically proven symp-
tomatic IGM.
exact cause remains unproven and the diagno-
sis is usually made by means of exclusion. The
clinical and radiologic findings of IGM are noted
to frequently overlap with those of breast cancer
and several benign inflammatory breast condi-
tions and thus can often lead to misdiagnosis and
delayed treatment. A standard diagnostic proto-
col has not yet been established. However, a diag-
nostic and management algorithm that includes
suggestions from the available literature (11)
combined with the current management strategy
used at our home institution (University of Texas
Health at San Antonio) is proposed herein.
Before the 1980s, most patients with IGM were
treated with wide surgical excision exclusively.
However, conservative therapy with oral steroids
or imaging surveillance is currently being endorsed
as a first-line treatment option, before surgical
consideration (12). Multiple case series have been
published since the 1980s, up to the year 2016.
These series include large cohort studies (one
involving 206 patients) from the Middle East and
Asia that include reports of different experiences
involving IGM with characteristic trimodality
imaging findings, usual clinical manifestations, and
evolving management options (4,5,13). We review
the clinical manifestations, major differential di-
agnoses, and histopathologic features of IGM. We
also describe current management strategies and
the critical roles of breast imaging and interven-
tional procedures in the setting of IGM.
Pathophysiologic Features
The cause of IGM remains unproven, and review
of the current literature reveals multiple theories
based on findings in small patient cohorts and/or
anecdotal observations. The most accepted the-
ory is that an initial insult to the ductal epithelial
cells in the breast causes a transition of luminal
secretions to the lobular breast stroma. This tran-
sition causes a local inflammatory response in the
connective tissue, with macrophage and lympho-
cyte migration to the region, and subsequently
a local granulomatous response (7). Multiple
inflammatory breast diseases that result from this
cascade of events, including periductal mastitis
and mammary duct ectasia, have been pro-
posed to coexist in a pathologic spectrum. These
diseases have been conceptually grouped into
a body of entities referred to as the mammary
duct–associated inflammatory disease sequence
(MDAIDS) (14). Several precipitating factors
have been proposed and include pregnancy, lac-
tation, hyperprolactinemia, α1-antitrypsin defi-
ciency, oral contraceptive use, trauma, diabetes,
autoimmune disease, and smoking. However,
pregnancy, lactation, and hyperprolactinemia are
the only factors with well-established associations
with IGM. These factors are discussed briefly in
this review (8,9,15,16).
Demographic Features
IGM is a rare disease of the breast, and its true
prevalence is not well established (17). Baslaim et
al (10) conducted a 10-year retrospective study
in a tertiary medical center in Saudi Arabia. They
found that of 1106 women with benign breast
diseases (mastalgia, fibrocystic changes, fibroad-
enomas, ductal ectasia, simple cysts, and acute and
chronic inflammatory conditions), only 1.8% of
them represented histopathogically proven cases
of IGM. The majority of individuals with IGM are
female, with a few reported cases in males (17–19).
Anecdotal explanations for inflammatory ductal or
lobulocentric disease in males include gynecomas-
tia as a predisposing factor and trauma, smoking,
and autoimmunity as exacerbating agents (20,21).
The findings in multiple studies (5,6,11,13,22–24)
indicate that IGM is almost always seen in women
of childbearing age. In a retrospective study (5)
involving 206 patients with histopathologically
confirmed IGM, the patients had a mean age of
32 years, with similar mean ages, up to 34 years,
reported in smaller studies (11,23,24). However,
the age range of affected patients can vary from late
childhood to the late postmenopausal period, with
some reports of patients as young as 11 years and
as old as 83 years (3,8).
There is a strong association between IGM
and history of pregnancy and lactation, with most
patients reporting having a pregnancy within
5 years before the diagnosis (6,25,26) or being
diagnosed with IGM within 2 months to 20 years
332  March-April 2018 radiographics.rsna.org

after a pregnancy (7,8). The majority of patients

report a nursing duration of 3–36 months and
having symptomatic IGM 6 months to 2 years
after the cessation of breast-feeding (7,8,17).
Altintoprak et al (7) assessed nursing habits in a
small sample of patients with IGM (N = 26) and
reported that a small subset of patients (15%)
who lactated exclusively from one breast devel-
oped IGM in the nonlactating breast; however,
no statistically significant associations could be
established. If a nonlactating breast in a lactating
female is identified as an important risk factor
for IGM development, this would further sup-
port the concept of an MDAIDS, with stasis- and
ectasia-causing ductal injury as a common risk
factor of IGM that is also seen with puerperal
(lactational) mastitis, ductal ectasia, and periduc-
tal mastitis (14).
The manifestation of IGM during pregnancy
and lactation is uncommon (5,8,10,24,26). Rare
cases of biopsy-proven IGM in nulliparous and
nonreproductive patients have been reported
(5,8), and some of these cases have been at-
tributed to increased levels of prolactin (9,27).
Hyperprolactinemic states, either drug induced
or caused by an intracranial lesion, have been
associated with the development of IGM. This as-
sociation lends support for the use of bromocrip-
tine as an optional conservative second-line
therapy for patients with IGM. The resolution
of breast symptoms after the use of antiprolac-
tinemic drugs and the correlation of markedly
high prolactin levels with treatment-resistant and
highly symptomatic IGM support such an asso-
ciation (9,28). IGM has been reported in nul-
liparous patients who (a) had hyperprolactinemia
related to the use of antipsychotic agents (ie,
risperidone), phenothiazine, or metoclopramide;
(b) experienced blunt trauma; and/or (c) had
pituitary neoplasms (9,27).
IGM is encountered worldwide in individu-
Figure 1.  Drawings of external anterior oblique view (a)
als of all races, without a well-established ethnic and sagittal view at the nipple axis (b) illustrate typical clinical
predisposition (17,29). However, some authors manifestations of IGM in the left breast. A peripheral inflamed
have reported an association with nonwhite indi- terminal ductal lobular unit (TDLU) with focal masslike proper-
viduals—Asian, Hispanic, and Middle Eastern ties is shown in b. The overlying focal skin erythema shown in a
is seen less commonly, in about one-third of patients. However,
persons in particular (6,17,23,29). In one study, it would be seen adjacent to the palpable finding and occu-
Al-Khaffaf et al (6) found that in an area with pying not more than one-third of the breast skin. A concur-
11.8% of the population reportedly comprising rent abscess with a draining cutaneous sinus tract is relatively
nonwhite persons, eight (44%) of 18 individuals common. Clinically palpable axillary lymph nodes, extensive
inflammatory skin changes, and nipple retraction are uncom-
with IGM were nonwhite persons. This finding mon with IGM.
supports the possibility of a race-related predis-
posing factor of IGM. In an Australian study in-
volving 17 patients with IGM, Skandarajah and white. However, these results were similar to the
Marley (29) found that 46.9% of patients were racial demographic distribution of patients who
nonwhite individuals: 17.6% of these patients visited that hospital, and no statistically signifi-
were of Mediterranean ancestry; 17.6%, of cant association could be made.
Chinese ancestry; and 11.7%, of subcontinental An association between IGM and Hispanic
ancestry. Fifty-three percent of the patients were ancestry in the continental United States was
RG  •  Volume 38  Number 2 Pluguez-Turull et al  333
studied in a case control study (30) involving 49
patients with IGM and 196 age-matched ran-
domly selected patients from a public hospital
breast clinic in Chicago, Illinois. The patients
in the IGM group were found to be statistically
significantly more likely to be Hispanic than the
patients in the control group, with an odds ratio
of 3.0 (95% confidence interval: 1.42, 6.24;
P =.003) (30,31). However, important biases
existed in that study: The researchers predomi-
nantly surveyed a low socioeconomic population
with diverse ethnicities, both patient groups were
predominantly Hispanic, and individuals of no
other ethnicities were evaluated. Other ethnic
associations have been made anecdotally, as
many of the reported cases have come from Asia,
Turkey, Jordan, and Iran (10,23).
Clinical Manifestations
The results of several studies (5,10,11,17,32) in-
dicate that the most common clinical manifesta-
tion of IGM is a tender palpable unilateral breast
mass of variable size (1–20 cm). Synchronous
bilateral breast findings were estimated to be seen
in 1% of cases in a study by Aghajanzadeh et al
(5) involving 206 patients and in up to 18% of
cases in a study by Gautier et al (17) involving 11
patients. Aghajanzadeh et al (5) also described
a tender palpable mass with skin erythema and
edema (in 11%–31% cases) and isolated skin
induration (in 20% of cases) as less common
clinical manifestations (Fig 1). Investigators in
a smaller retrospective study involving 20 cases
(10) reported peau d’orange skin changes in 40%
of patients and asymmetric breast heaviness or
enlargement in approximately 20% of patients.
These findings are also seen with inflammatory
breast cancer (IBC).
The nipple is seldom involved; however, nipple
retraction, ulceration, and secretions have been
documented. Investigators in a study involving
206 patients with IGM (5) reported nipple secre-
tions in 12% of cases and nipple-areola complex
ulcerations in 16% of cases. Axillary lymphade-
nopathy is only occasionally detected at physical
examination—in approximately 28% of patients
with IGM (5,13). This finding is often more con-
cerning with regard to possible breast malignancy
with nodal metastatic disease (32).
IGM may manifest with abscess formation,
with or without draining skin sinuses, at a vari-
able prevalence of 6.6%–54.0% (4,5,26,33). A
cutaneous fistula can develop as a complication
of prior percutaneous biopsy or aspiration. These
abcesses are typically aspirated for microbiologic
analysis. Data in the current literature indicate that
most IGM-associated abscesses are sterile, with-
out bacterial growth at aspirate culture analysis,
suggesting a noninfectious inflammatory process
(4,33). It is important to note that the presence of
an abscess at clinical examination and/or imaging
should not preclude tissue biopsy if the clinical
course suggests possible malignancy, as IBC also
can manifest with fluid collections (17).
Core-Needle Biopsy with or without
Fine-Needle Aspiration
Ultrasonographically (US)-guided fine-needle
aspiration (FNA) and core-needle biopsy, with or
without aspiration of fluid collections, are breast
imaging interventions commonly performed in
patients with IGM. FNA can be used to render
a histopathology-based diagnosis; however, its
usefulness and reliability have been widely debated
(5,24,34). Although FNA may be helpful initially
for distinguishing an inflammatory breast process
from a malignancy, a definitive histopathology-
based diagnosis eventually should be rendered by
using core-needle, vacuum-assisted, or excisional
biopsy (11,13,17,35). In a study of 14 cases of
IGM, Martínez-Parra et al (34) concluded that
FNA is not sufficient for confidently characteriz-
ing the aspirates of IGM and distinguishing them
from the aspirates of other types of granulomatous
diseases. Similarly, in a study conducted by Kok
and Telisinghe (24) involving 23 patients with
IGM, in which FNA was initially performed and
the diagnosis was confirmed with core-needle and/
or excisional biopsy, FNA facilitated a diagnosis
in only four of 24 IGM lesions. In a study involv-
ing 206 patients with IGM, Aghajanzadeh et al
(5) found that FNA was diagnostic in 39% of the
patients, while core-needle biopsy was diagnostic
in 94.5% of them.
Given the poor diagnostic rate associated with
FNA, several authors believe that it is an unnec-
essary intervention in the management of IGM,
unless it is performed for drainage of associated
fluid collections (5). It is conceivable that having
a cytologist on site could improve the diagnostic
effectiveness of FNA biopsy, with possible im-
mediate conversion to core-needle biopsy when
needed.
Core-needle biopsy has a well-established
role in the diagnosis of IGM, with up to
94%–100% accuracy reported in several studies
(5,13,17,22,35). It also enables more extensive
testing to be performed in cases of infection,
malignancy, and other noninfectious inflamma-
tory breast diseases. It remains uncertain whether
performing core-needle biopsy markedly exacer-
bates the inflammatory changes in quiescent or
mildly symptomatic IGM. Several authors have
endorsed the second-line use of vacuum-assisted
biopsy when (a) no definite diagnosis can be
established with core-needle biopsy, (b) there is
334  March-April 2018 radiographics.rsna.org

Table 1: Reported Imaging Findings of IGM

Finding Prevalence (%)

Mammography
  Focal or global asymmetry 36–75
  Irregular focal mass 11–67
  Normal findings 8–45
  Axillary adenopathy 15–18
  Skin thickening with edema or trabecular thickening 5–21
  Asymmetrically increased breast density 4.5–17.0
  Architectural distortion 9
  Circumscribed mass 9
 Calcifications Very rare
US
  Irregular hypoechoic mass with tubular extensions 40–100
  Axillary adenopathy 28–60
  Circumscribed hypoechoic mass 25–52
  Skin thickening and edema 17–60
  Abscess and/or sinus tract 6.6–54.0
Heterogeneous hypoechoic mass (or confluent masses) with 6.6–33.0
indistinct, lobulated, or angular margins
Parenchymal distortion with or without acoustic shadow- 4.0–26.7
ing, no discrete mass
  Normal findings 3.4–20.0
  Heterogeneous parenchyma or parenchymal edema 10–13
Multiparametric MR imaging
  T2 hyperintensity (edema) of breast stroma Majority
Rim-enhancing lesions (microabscesses) or heterogeneously 71–86
enhancing masses, with or without NME
  Segmental or regional NME 30–80
Contrast enhancement with variable kinetic properties:
   Type I 38.0–82.7
   Type II 13.8–40.0
  T2-hypointense enhancing mass with irregular margins 20
Note.—MR = magnetic resonance, NME = nonmass enhancement.

a discrepancy between the radiologic and histo-
pathologic findings, or (c) a target for biopsy can
be seen at mammography only. Vacuum-assisted
biopsy techniques enable larger volume sampling
with improved diagnostic sensitivity and specific-
ity for breast diseases (17,36).
In the majority of patients, there is no indica-
tion for or advantage to performing open-exci-
sion biopsy, as it can lead to substantial scarring,
breast asymmetry or deformity, and nonhealing
ulcers—the latter of which can eventually lead
to sinus tract formation (5,11). Aghajanzadeh et
al (5) found that 19 (44%) of 43 patients who
underwent surgical biopsy eventually developed
ulcers and/or sinus tracts, which were not com-
monly observed in the patients who underwent
FNA or core-needle biopsy. Therefore, excision
biopsy is usually reserved for cases in which
there is a discordance between the radiologic
findings and core-needle or vacuum-assisted
biopsy findings (10,13). A surgeon or derma-
tologist may perform a skin-punch biopsy in
patients who have extensive skin inflammation
when there is a high suspicion for IBC, particu-
larly when there is no underlying focal breast
abnormality (37). However, skin-punch biopsy
is not required for the diagnosis of IBC when
there is histopathologic evidence of ipsilateral
breast malignancy (38).
Imaging Overview and Strategy
At imaging, granulomatous mastitis can manifest
with a variety of nonspecific appearances (Table
1), which often mimic the appearances of ma-
lignancy. Much of the variability in the imaging
appearances of IGM might be related to varying
RG  •  Volume 38  Number 2 Pluguez-Turull et al  335
histopathologic features, including inflammatory
reaction, abscess, and fibrosis (39).
The current strategy for imaging IGM includes
the use of mammography and US, with further
imaging tailored to the patient and the imaging
findings. There is often a deviation from this strat-
egy for patients younger than 40 years who present
with a palpable abnormality, with mammography
omitted (11,39). The following imaging protocol
is endorsed in the current American College of
Radiology Appropriateness Criteria: initial evalu-
ation of palpable breast abnormalities with US for
patients younger than 30 years and with mam-
mography for patients older than 40 years, and
discretionary initial evaluation of palpable breast
abnormalities for patients aged 30–40 years (40).
However, Dursun et al (41) and other authors
support performing optional unilateral mammog-
raphy of the breast of concern in patients younger
than 40 years. In most other scenarios, bilateral
mammography is performed with standard cra-
niocaudal (CC) and mediolateral oblique (MLO)
views, with subsequent additional views obtained
as warranted (11). Targeted US with a high-
frequency linear probe is nearly always performed,
as the patient often presents with a palpable area,
focal pain, and/or focal skin changes (42). In
addition, US enables further characterization of
mammographic findings. The limitations of these
imaging modalities are largely related to pain,
which limits the patient’s tolerance to compression
or pressure from the US transducer, or edema,
which limits evaluation of the breast parenchyma.
While mammography and US are often suf-
ficient for imaging granulomatous mastitis, MR
imaging may offer additional benefits in the
evaluation of advanced, aggressive, or refractory
disease (32). MR imaging may be indicated when
US and mammographic evaluation is limited by
parenchymal or cutaneous edema and/or breast
density, or for identifying a biopsy target (13,17).
It may also be helpful in delineating the extent of
disease, including evaluation of the contralateral
breast. In addition, MR imaging may be useful for
evaluating possible residual disease after treat-
ment or for monitoring disease in patients who
are undergoing conservative treatment (17,32). It
is conceivable that MR imaging could be used in
patients with an intermediate to elevated (>15%
to 20%) lifetime risk of malignancy or docu-
mented high-risk breast lesions. Aside from these
indications, MR imaging might not yield addi-
tional information and probably will not be useful
for distinguishing mastitis from other mimick-
ing entities, including IBC (42,43). It is critical
that performing MR imaging, when indicated,
does not contribute to a delay in the diagnosis or
treatment.
The distribution of IGM is usually unilateral,
with bilateral cases reported less often, in 1%–9%
of cases (5,32,42,44). However, bilateral IGM
was reported in up to 18% of cases in the Gautier
et al study (17). The incidence of multicentric or
multifocal disease is not well documented in the
literature. Most lesions occur in the breast pe-
riphery, with other lesions occurring in a subare-
olar location or with diffuse involvement. Preva-
lences of peripheral, subareolar, and diffuse IGM
lesions of 50%, 25%, and 25%, respectively, have
been reported (11,13,26,44). An IGM lesion
can develop in any quadrant of the breast, with
no association with a particular quadrant noted
(8,24). There are multiple reports of a right-side
predominance, in 61%–69% of cases, that could
not be explained (5,8,13).
Typical Imaging Appearances of IGM
Mammographic Findings
The mammographic findings of IGM are non-
specific and varied. A review of the literature
reveals that the most frequently encountered
mammographic appearance is focal asymmetry
(Figs 2, 3) (11,13,17,32,39,41,45–47); however,
investigators in a few studies argue that the most
common manifestation at mammography is an
irregularly shaped or obscured mass (5,44) (Fig
4). Often, there is no mammographic finding—
especially in the setting of heterogeneously or
extremely dense breasts (Fig 5) (48). In several
studies, asymmetrically dense breast parenchyma
or global asymmetry has been noted as a less
common manifestation of granulomatous mastitis
(11,25) (Fig 6), and architectural distortion is
reported less frequently (41).
Additional mammographic findings include
axillary adenopathy and focal skin thickening
or edema, the reported frequencies of which
vary greatly (5,11,13,26,32,41). In contrast to
IBC, which characteristically involves more than
one-third of the breast skin, IGM rarely involves
the skin extensively (49,50) (Fig 7). IGM is not
usually associated with calcifications. Fazzio et al
(32) found calcifications to be a very rare mam-
mographic finding of granulomatous mastitis,
citing a single case that appeared as segmental
coarse heterogeneous calcifications. However,
there are case reports of symptomatic IGM
masking a synchronous malignancy at presenta-
tion, which is often seen as microcalcifications on
mammograms obtained after the breast symp-
toms have been resolved (51). These findings
support the utility of mammographic follow-up
after the resolution of breast pain or tenderness,
which often limits the initial mammographic
evaluation in women in whom it is indicated.
336  March-April 2018 radiographics.rsna.org

Figures 2, 3.  (2) IGM in a 37-year-old woman with a painful mass of 3 weeks’ duration in the outer region of the left breast.
(a, b) MLO (a) and CC (b) digital mammographic views show focal asymmetry (arrow) in the outer region of the left breast.
(c) Targeted US image at the 3-o’clock position in the left breast shows a heterogeneously hypoechoic parallel mass with hypoechoic
tubular extensions and surrounding inhomogeneous breast tissue. FN Rad = from nipple radial. Some degree of posterior shadowing
(not shown) was reported, and the mass was classified as a Breast Imaging Reporting and Data System (BI-RADS) category 4 lesion
at subsequent US-guided biopsy, which yielded benign breast parenchyma with chronic granulomatous inflammation. The patient
underwent conservative management. (3) IGM in a 32-year-old woman with a palpable left-breast lump and associated breast
tenderness. (a, b) CC (a) and MLO (b) digital mammographic views show focal asymmetry (arrow) in the inferior region of the left
breast, at the 4-o’clock to 8-o’clock axes. (c) Targeted US image at the 4-o’clock position shows a large irregular hypoechoic lesion
with tubular extensions, which is the most common US finding of IGM. A Rad = antiradial. This was classified as a BI-RADS category
4 lesion. Subsequent US-guided core-needle biopsy facilitated a diagnosis of granulomatous mastitis. The patient underwent surgical
consultation and conservative treatment.

US Findings commonly reported at presentation (13,41,44).

US findings of IGM also vary, with a large irregu-
lar hypoechoic parallel mass with tubular exten-
sions being the most frequently reported mani-
festation (5,11,13,26,32,39,44). This finding was
present at US in 59% of patients in a retrospec-
tive analysis of data from more than 200 patients
(5). A similar description of confluent lesions or
collections with tubular extensions also has been
described as the more commonly seen finding
(Figs 2, 3) (17,45). The tubular extensions dem-
onstrate how IGM insinuates around the breast
lobules rather than destroys them (17). A circum-
scribed hypoechoic parallel mass also has been
In other retrospective analyses (11,25,32,41,48),
a mass lesion that often demonstrates angular,
lobulated, or indistinct margins (Fig 8) has been
reported as the most common finding. Posterior
acoustic features vary greatly, with both posterior
acoustic enhancement and shadowing having
been described (5,17,32). Lesions are nearly al-
ways parallel in orientation (44). Doppler US im-
ages show the lesion—and often the surrounding
tissue—to be hypervascular (25,32,48). With ad-
vanced disease, fluid collections or abscesses may
be present, with reported prevalences ranging
from 6.6% to 54.0% (25,26,32,33,39,41). Ad-
RG  •  Volume 38  Number 2 Pluguez-Turull et al  337

Figure 4.  IGM appearing as an obscured mass in a 52-year-old woman with a palpable retroareolar
left-breast lump discovered 4 days earlier. (a, b) CC (a) and MLO (b) mammographic views of the
left breast show a round dense mass (arrow) with obscured margins in the palpable region of concern.
(c) Targeted US image of the left breast shows a round, heterogeneously hypoechoic mass with indistinct
margins and minimal ductal dilatation. Rad = radial. The mass was classified as a BI-RADS category 4
lesion at subsequent US-guided core-needle biopsy, which yielded granulomatous mastitis. The patient
underwent imaging surveillance.

ditional US appearances include heterogeneous
breast parenchyma, a circumscribed hypoechoic
mass, and parenchymal distortion with acoustic
shadowing but without a discrete mass (Fig 9)
(11,39,41). In a smaller subset of IGM cases,
there are no relevant US findings that correlate
with the positive mammogram findings (Fig 10)
(13,46). Ancillary US findings include skin thick-
ening and edema, subcutaneous fat obliteration,
and axillary adenopathy (smooth reactive corti-
cal thickening with preserved fatty hilum), the
reported frequencies of which also vary widely
(5,11,17,25,32,44). However, it has been noted
that the imaging findings of clinically undetected
axillary adenopathy and skin thickening are usu-
ally seen more frequently on US images than on
concurrently obtained mammograms (11).
MR Imaging Findings
The MR imaging appearances of IGM, as with
the US and mammographic appearances, vary
greatly; however, the sensitivity of this modal-
ity is high, with no reported negative correlates
(13,17,26,32,39,41). Although MR imaging is
used less frequently for the evaluation of IGM,
several reports in the literature (13,17,39,41)
involve sample sizes of nine, 20, 29, and 36 patients
and thus provide reviews of the most frequently
encountered MR imaging appearances to be
discerned. A heterogeneously enhancing mass (or
masses) or rim-enhancing lesions are the most
commonly described findings seen on MR im-
ages, which may also show associated segmental or
regional nonmass enhancement (NME) (Figs 11,
12) (13,25,32,39,41). Some small lesions demon-
strating confluency or well-defined borders, T2 hy-
perintensity, and rim enhancement at MR imaging
have been presumed to represent microabscesses
(13,17,32). NME without an accompanying mass
also has been commonly seen (30%–80%), and
in general, the NME with IGM has more com-
monly demonstrated a segmental rather than
regional distribution. Diffuse NME is rarely seen
(13,26,39,41). Oztekin et al (13) reported the pres-
ence of a frank abscess with associated NME in 25
(86%) of 29 patients with a diagnosis of granulo-
matous mastitis, which manifested solely as NME
in the remaining patients. However, frank abscesses
have not been reported by most other authors.
Yildiz et al (26) encountered a less common
MR imaging finding of IGM, a T2-hypointense
mass with irregular margins, in six (20%) of 30 pa-
tients. The majority (80%) of these patients were
found to have a T2-hyperintense enhancing mass
with or without NME, a finding that may reflect
varying degrees of fibrosis. In general, the mar-
gins and shapes of these masses vary greatly and
are described as ill defined or well circumscribed
(margin), and round, oval, or irregular (shape)
(25,39,41). Restricted diffusion and T2 hyperin-
tensity (representing edema) of the affected breast
parenchyma are seen in the majority of cases (32).
Additional MR imaging findings include axillary
lymphadenopathy, nipple and/or skin thickening,
nipple retraction, sinus tracts, and parenchymal
distortion (17,32,41). The enhancement kinetic
338  March-April 2018 radiographics.rsna.org

Figure 5.  IGM appearing as a mammographically occult palpable left-breast mass with subjective
growth during the past 3–4 months in a 30-year-old woman. (a, b) CC (a) and MLO (b) mammographic
views show an extremely dense left breast with no focal abnormality. (c) Targeted US image obtained at
the 12-o’clock to 2-o’clock axes shows a hypoechoic mass with indistinct margins. FN Arad = from nipple
antiradial. The mass was classified as a BI-RADS category 4 lesion, with subsequent US-guided core-needle
biopsy yielding benign granulomatous inflammation with a sterile abscess. Abscess aspiration was at-
tempted at biopsy. The patient underwent conservative management and surveillance.

Figure 6.  IGM appearing as asymmetrically increased left

breast density at mammography in a 30-year-old woman who
reported having hardness and pain in the lower outer region of
the breast. Physical examination revealed skin erythema and ten-
derness. She also reported having a fever, which improved mildly
after she took ibuprofen, and multiple previous similar episodes,
which improved after antibiotic treatment and recurred at the
same site. (a, b) Right-breast MLO (a) and left-breast MLO (b)
mammographic views show heterogeneously dense breasts, with
asymmetrically increased left breast density or global asymmetry,
which appears to be more pronounced inferiorly. (c) Targeted
US image at the 4-o’clock to 6-o’clock position in the left breast
shows a hypoechoic appearance of the fibroglandular tissue,
without a discrete mass, cyst, or drainable fluid collection. FN
Rad = from nipple radial. The nonresponsiveness to antibiotics,
the patient’s age, and the imaging features favored a diagnosis of
atypical infective mastitis or granulomatous mastitis; a diagnosis
of malignancy was less likely. The finding was therefore classi-
fied as BI-RADS category 4. The US-guided core-needle biopsy
specimen revealed chronic granulomatous inflammation and was
negative for fungi and acid-fast organisms.

properties are nonspecific, although the majority

of cases involve progressive or plateau patterns of
enhancement (13,32,39,41). Plateau or washout
patterns of lesion enhancement are reported in
fewer studies (25). In other studies, ring-enhancing
lesions often display washout while NME involves
progressive enhancement (17). Owing to this
variability in enhancement kinetic properties, MR
imaging cannot be used to reliably distinguish IBC
from granulomatous mastitis (42,47).

Other Imaging Modalities

There are very few literature reports on the com- be part of the workup for IGM. In addition, data
puted tomographic (CT) appearances of granulo- regarding the appearance of granulomatous mas-
matous mastitis, and CT is not recommended to titis at breast tomosynthesis are lacking; however,
RG  •  Volume 38  Number 2 Pluguez-Turull et al  339

Figure 7.  IBC appearing as an abscess in a 28-year-old woman who presented with a tender, rapidly growing
right-breast mass with erythema. Multiple subsequent antibiotic treatment trials were unsuccessful, and abscess
aspirations yielded sterile cultures and no malignancy. The abscess was suspected to represent IGM. (a) Tar-
geted US image obtained at the 6-o’clock to 10-o’clock position in the right breast for initial evaluation shows
a large complex mass (arrowhead) with solid and cystic components and internal vascularity. The mass was
believed to represent surrounding inflammation and/or granulation tissue associated with abscess formation and
was classified as a BI-RADS category 3 lesion. (b) Three weeks later, after empiric antibiotic therapy, the patient’s
symptoms remained and she underwent repeat US of the outer lower quadrant of the breast, which depicted
similar findings, prompting aspiration, which yielded a sterile abscess without malignant cells. The abscess was
classified as a BI-RADS category 3 lesion. (c) After 8 weeks and two completed antibiotic treatment trials, the pa-
tient’s symptoms persisted and the diagnosis of IGM was considered. At this time, MLO mammography was per-
formed and depicted a dominant, circumscribed round right-breast mass associated with architectural distortion
and surrounding trabecular thickening (thick arrow). Extensive skin changes were less readily observed owing to
artifact (thin arrow). (d) Targeted US image obtained at the 6-o’clock position shows an oval hypoechoic mass
(arrowhead) with indistinct posterior margins and prominent internal vascularity. The mass was classified as a
BI-RADS category 5 lesion, with US-guided core-needle biopsy yielding high-grade invasive ductal carcinoma
with a Ki67 cell proliferation rate of 99%.

abscess formation and the related possibility for in-

tervention, (d) evaluate the stability of or interval
change in the lesion(s), (e) evaluate the treatment
response, and (f) identify metachronous disease
and local recurrence at imaging surveillance.

it is conceivable that this modality may be a part
of the evaluation. Similarly, the role of molecular
breast imaging in the evaluation of IGM has not
been established.
Imaging in the Evaluation of
Histopathologically Confirmed IGM
Regardless of the imaging strategy used to diag-
nose IGM, the roles of the breast imager in the
surveillance, presurgical, and posttreatment evalu-
ation of confirmed IGM are to (a) establish the
multiplicity and location of IGM lesions,
(b) document the size of the lesion(s), (c) identify
Differential Diagnoses
The main differential diagnoses for IGM include
malignancy and other benign inflammatory con-
ditions of the breast (Table 2). Mammography
and US performed in cases of suspected IGM, in
conjunction with clinical history assessment, are
most helpful for evaluating features that are more
likely to be associated with a malignancy rather
than a benign process that does not require im-
mediate intervention.
Inflammatory Breast Cancer
The most worrisome diagnosis to exclude is IBC,
an aggressive form of breast cancer that involves
lymphovascular invasion and often mimics other
inflammatory breast diseases clinically and
radiologically (Fig 13) (8). There is an oncologic
consensus regarding the clinical criteria that are
important for the diagnosis of IBC, as both a
tissue specimen–based diagnosis of malignancy
and clinical evidence of inflammatory disease
are required to confirm the diagnosis of IBC.
Signs and symptoms that strongly suggest IBC
340  March-April 2018 radiographics.rsna.org

Figure 8.  IGM appearing as a mammographically occult palpable mass in a 26-year-old woman with
a history of a tender enlarging right-breast mass that improved clinically after antibiotic treatment initia-
tion but then recurred within a few days. (a, b) CC (a) and MLO (b) mammographic views of the right
breast show no substantial abnormality. (c) Targeted US image obtained at the 10-o’clock position in the
right breast shows an oval parallel heterogeneously hypoechoic mass with indistinct margins and inter-
nal and rim vascularity corresponding to the palpable area of concern. Arad = antiradial. This mass was
categorized as a BI-RADS category 4 lesion, with US-guided core-needle biopsy revealing granulomatous
mastitis. The patient underwent conservative treatment.

include erythema occupying at least one-third of
the breast, rapid onset of skin edema and/or peau
d’orange, and/or a warm breast with or without
an underlying palpable mass. The onset of signs
and symptoms characteristically occurs within 6
months or less of the initial presentation (52).
The presence of clinically palpable axillary
lymph nodes and unilateral breast enlargement
increases suspicion, as these are common initial
clinical manifestations of IBC and are seldom
seen with IGM (5,11,53). Patient age is a useful
factor to consider when evaluating for possible
IGM versus IBC. In a small retrospective study,
Wang et al (54) compared the findings in pa-
tients with breast cancer with those in patients
who had benign inflammatory diseases, includ-
ing IGM, and found that the mean age of the
patients with cancer (55.4 years ± 13.9 [stan-
dard deviation]) usually was older than that of
those with benign inflammatory diseases (44.5
years ± 11.3). It could be argued that the mean
age of persons with isolated IGM (32–34 years)
further supports the differentiation of these two
entities on the basis of this factor.
At mammography of IBC, the most common
findings are extensive skin edema and trabecular
thickening, which are less common with IGM.
An accompanying breast mass, asymmetry, and/
or architectural distortion can be seen with both
IBC and IGM. At US, extensive skin thicken-
ing and breast edema with dilated lymphatics
are more characteristic of IBC; heterogeneous
parenchyma, axillary adenopathy, and irregular or
conglomerate masses are less discriminatory. At
MR imaging, extensive skin thickening with asso-
ciated enhancement; breast enlargement; axillary
adenopathy; increased breast signal intensity on
T2-weighted images, which may extend to the
chest wall; and a rapidly enhancing breast mass
or abnormal parenchymal enhancement are more
suggestive of IBC (53,54).
Infective Mastitis
Infective mastitis is a more typical diagnosis
to exclude in the setting of suspected IGM, as
it is the most common cause of inflammatory
breast disease in women of childbearing age (Fig
14). Infective mastitis can be seen in lactating
and nonlactating women of all ages. In a study
conducted by Kamal et al (55) involving 197
female patients aged 14–67 years (average age,
39.8 years) who had a clinical or histopathologic
analysis–based diagnosis of mastitis, 67% of the
cases were attributed to infection. This group also
found infective mastitis to be very common dur-
ing the childbearing period and during lactation,
with 63.6% of patients being younger than 40
years and 37.9% of them lactating (55).
The mammographic appearance of simple
mastitis includes trabecular thickening and focal
RG  •  Volume 38  Number 2 Pluguez-Turull et al  341

Figure 9.  IGM in a 34-year-old woman who had a very tender mass in the left breast about 1 week previously. She
reported having started a 10-day course of antibiotics 7 days earlier, with mild improvement. (a–c) IGM appeared
as focal asymmetry with nipple inversion at mamography (a, b) and as heterogeneous breast tissue with areas of
mild posterior shadowing at US (c). CC (a) and MLO (b) mammographic views of the left breast show focal asym-
metry (thick arrow in b) in the left lower inner quadrant, with retroareolar architectural distortion, nipple inversion
(thin arrow in b), and mild skin thickening (arrowhead in b). Targeted US image at the 8-o’clock position in the
left breast (c) shows heterogeneously hypoechoic breast tissue. The findings were classified as representing a BI-
RADS category 3 lesion, which was suspected of being infective mastitis with low suspicion for malignancy, and the
patient was allowed to finish the antibiotic course. After 35 days, she returned to the treatment facility with increas-
ing left-breast tenderness and skin erythema after having undergone multiple courses of antibiotics. Rad = radial.
(d) Targeted US image at the 9-o’clock position in the left breast now shows an oval parallel hypoechoic mass that
has internal vascularity (arrow) and some internal fluctuance, with surrounding tissue edema. Arad = antiradial, FN =
from nipple. The findings were classified as representing a BI-RADS category 4 lesion, with subsequent aspiration
and US-guided core-needle biopsy yielding a sterile abscess in the setting of granulomatous mastitis. The lesion
eventually became complicated by a cutaneous fistula in the medial region of the left breast.

Figure 10.  IGM seen as an ir-

regular subareolar mass at mam-
mography, with no correlative US
finding, in a 62-year-old woman
with left-breast pain and a spon-
taneous yellowish nipple discharge
of 2 weeks’ duration. Right-breast
MLO (a), left-breast MLO (b), and
left-breast CC (c) mammographic
views show an equal-density irreg-
ular mass (arrow in b and c) with
irregular margins in the subareolar
region of the left breast. The pres-
ence of invasive lobular carcinoma
or invasive ductal carcinoma was
considered, and a classification
of BI-RADS category 4 lesion was
recommended. Subsequent ste-
reotactic vacuum-assisted biopsy
revealed granulomatous mastitis,
and the patient underwent conser-
vative treatment.
342  March-April 2018 radiographics.rsna.org

Figure 11.  IGM appearing as asymmetrically increased breast density at mammography and as an irregular hypoechoic mass with
tubular extensions at US in a 24-year-old woman who presented (at an outside institution) with a hard tender breast mass in the left
breast. (a–c) Right-breast MLO (a), left-breast MLO (b), and left-breast CC (c) mammograms show asymmetrically increased density
of the left breast, as compared with the density of the right breast, without focal findings. (d) Left-breast US image obtained in the
periareolar region shows a large heterogeneous hypoechoic masslike area with tubular extension. Short-term imaging follow-up
and a lesion classification of BI-RADS category 3 were recommended. One month after the initial evaluation, the patient presented
at our institution with persistent symptoms, having completed two 10-day courses of antibiotics. Physical examination revealed a
mildly fixed mass at the 11-o’clock position in the left breast. (e–g) Axial T1-weighted fat-saturated (e), gadolinium-based contrast
material–enhanced T1-weighted fat-saturated (f), and T2-weighted fat-saturated (g) breast MR images were obtained and showed
a large heterogeneously enhancing retroareolar mass (arrowhead in f) with increased signal intensity (circle in g) throughout the left
breast at T2-weighted imaging. Kinetic curves (not shown) demonstrated a type II (plateau) morphology. The mass was classified as
a BI-RADS category 4 lesion, with US-guided biopsy revealing granulomatous mastitis. After surgical consultation, the patient chose
to undergo surgical excision of the lesion.

or global asymmetry with or without skin thicken-

ing and axillary lymphadenopathy. Heterogeneous
breast tissue and skin or breast edema can be
seen at US. A superimposed abscess is common
in cases of infective mastitis, with characteristic
circumscribed or obscured oval to round masses
at mammography and anechoic to hypoechoic par-
allel collections with debris or complex heteroge-
neously hypoechoic avascular internal components
and rim hypervascularity at US.
Lactational (puerperal) mastitis is a subtype of
infective mastitis that occurs in lactating women
who develop a superimposed infection or abscess
in stagnant physiologic breast secretions. Mam-
mography typically is used to identify a fluid
collection, with subsequent drainage and cultures ces, Corynebacterium kroppenstedtii, and resistant
revealing common skin bacteria such as S aureus, bacterial strains (32,58). By definition, granulo-
Staphylococcus albus, or Streptococcus species matous mastitis should yield negative cultures,
(37,56,57). Atypical causes of infective mastitis although an association with Corynebacterium
also should be considered in patients who have infection has been proposed (16). We found
infections that are nonresponsive to empirical an- no comparative incidence analyses of atypical
tibiotic therapy. Usual causes of these infections sources of infective mastitis to determine the
include Mycobacterium, Echinococcus, Actinomy- most common agent. Aside from multiple cases
RG  •  Volume 38  Number 2 Pluguez-Turull et al  343

Figure 12.  Severe IGM in a 46-year-old woman who presented with pain, edema, skin erythema, skin ulceration, peau
d’orange, fevers, and a palpable mass in the left breast; these symptoms had been present for 2 weeks. Her symptoms
improved mildly after 4 days of oral antibiotics. (a, b) Initially obtained MLO (a) and CC (b) mammographic views of the
left breast show global focal asymmetry (thick arrow) in the left upper outer region, with nipple inversion (arrowhead in
a) and trabecular and skin thickening (thin arrow in b). (c) Targeted US image of the upper outer region of the left breast
shows a dominant 2.7-cm irregular complex fluid collection with internal septa (arrow) in the subareolar region at the
12-o’clock position. A RAD = antiradial. There were also two left axillary lymph nodes with mild cortical thickening and pre-
served fatty hila (not shown). Additional smaller complex fluid collections were seen at the 5-o’clock to 6-o’clock position
(not shown). Physical examination revealed stable left-nipple inversion and right-nipple retraction, which, according to the
patient, had been present for some years, as well as skin erythema and a 1-cm superficial skin ulcer at the 3-o’clock posi-
tion, approximately 5 cm from the nipple, on the left breast. A classification of BI-RADS category 3 was recommended, with
interval follow-up imaging after the completion of a 2-week course of antibiotics. (d–g) Progression of symptoms during
the follow-up period prompted a request for additional imaging, and axial T1-weighted fat-saturated (d), gadolinium-en-
hanced T1-weighted fat-saturated (e), and T2-weighted fat-saturated (f, g) breast MR images were obtained and showed
an asymmetrically enlarged left breast with a T2-hyperintense, heterogeneously enhancing (type I and type II kinetics not
shown) periareolar mass (* in f) with irregular margins extending to the skin (oval outline in e), multiple rim-enhancing
fat-fluid collections (arrow in f), regional heterogeneous NME (* in e), diffuse breast edema, skin thickening, and left-nipple
retraction (* in g). Type I NME (arrow in e) and multiple oil cysts in the right breast also were present. Excisional left-breast
biopsy revealed acute and chronic inflammation with abundant granulation tissue, which was negative for microorganisms
at Gomori methenamine-silver and Fite acid-fast staining and without cancer cells. The patient underwent complete left-
breast mastectomy with right-breast surveillance.
344  March-April 2018 radiographics.rsna.org

Table 2: Common Clinical and Imaging-based Differential Diagnoses for IGM

Clinical Histopathologic
Diagnosis Demographics Manifestations Imaging Findings Features
IGM Mainly affects pre- Palpable mass Mammography: focal asymmetric Lobulocentric
menopausal and Mastalgia with or density or irregular mass, tra- noncaseating
parous women without mild fo- becular and skin thickening granulomas
(after nursing cal skin erythe- US: irregular hypoechoic mass with Negative microbial
period) ma or draining hypoechoic tubular extensions staining and cul-
sinus MR imaging: heterogeneous en- ture results
History of failed hancing T2-hyperintense mass
antibiotic treat- and/or rim-enhancing lesions
ments with NME
IBC Mainly affects Skin erythema in at Mammography: skin and tra- Most often invasive
older women least one-third of becular thickening, asymmetric ductal carcinoma
(average age, 58 the breast increased breast density with that is poorly dif-
years, as com- Peau d’orange or without focal asymmetry, ferentiated, with
pared with 33 Asymmetric breast irregularly shaped mass, axillary dermal lympho-
years for IGM engorgement adenopathy vascular invasion
group) Onset to manifesta- US: extensive skin thickening and No inflammation
Higher prevalence tion of symp- breast edema, dilated lymphat-
in African- toms, less than 3 ics, axillary adenopathy, hetero-
American months geneous parenchyma with or
individuals Axillary adenopa- without suspicious or conglom-
thy in approxi- erate masses
mately 50%– MR imaging: breast and chest
85% of cases wall edema, streaky T2 hyper-
intensity, dilated lymphatics,
skin enhancement, contiguous
or coalescent irregular breast
masses with rapid enhancement
and washout kinetics (type III)
Infective Common in fe- Noncyclical breast Mammography (often not per- Abundant leuko-
mastitis males of repro- pain and/or ten- formed): trabecular and skin cytes
ductive age, but derness thickening, asymmetric in- Positive microbial
seen in persons Erythema creased breast density staining and cul-
of all ages Fever with or with- US: diffuse or focal skin thicken- ture results, with
out abscess ing, inhomogeneous breast Staphylococcus
Clinical unre- tissue with or without irregular and Streptococcus
sponsiveness to hypoechoic mass (with or with- bacteria often
empiric antibiot- out fluid collection) (particu- seen
ics in the pres- larly lactation mastitis) Inspissated secre-
ence of positive tions
microbial stains Atypical organisms
and/or cultures for which ad-
suggests an ditional staining
atypical or resis- is required for
tant organism identification
may be seen
Tuberculous Seen in endemic Palpable breast Mammography: findings similar to Caseating granu-
mastitis areas, high-risk mass those of infectious mastitis lomas
populations, Axillary lymphade- US: heterogeneous hypoechoic ir- Positive acid-fast
and persons nopathy regular mass, axillary lymphade- or Fite staining
with a history of Unilateral involve- nopathy with or without fluid results
pulmonary tu- ment collections
berculosis (50% Less mastalgia
of cases) compared with
the mastalgia
occurring with
IGM (continues)
RG  •  Volume 38  Number 2 Pluguez-Turull et al  345

Table 2: Common Clinical and Imaging-based Differential Diagnoses for IGM (continued)

Clinical Histopathologic
Diagnosis Demographics Manifestations Imaging Findings Features
Mammary Mainly affects Often incidental Mammography: tubular or branch- Dilated ducts with
duct ectasia perimenopausal Subareolar breast ing retroareolar structures with luminal, periduc-
and postmeno- mass with or thick rodlike (secretory) calcifi- tal, and stromal
pausal women without noncyc- cations lipid-laden histio-
lical breast pain US: dilated subareolar ducts, thick cytes
Unilateral or bilat- walls, anechoic fluid collections Periductal fibrosis
eral with debris (with or without in- Calcifications in
Nipple involvement traductal mass or filling defects) duct lumen or
and nonbloody MR imaging: retroareolar T2- wall
nipple discharge bright tubular structures
are common
Diabetic mas- Affects longtime Hard palpable Mammography: ill-defined, dense, No well-defined
topathy insulin-depen- mass(es) noncalcified mass(es) or asym- mass
dent females, Nontender metric densities Lymphocytic
persons with a Usually multiple US: irregular hypoechoic mass, infiltrates around
history of auto- and bilateral strong posterior acoustic shad- ducts
immune or en- owing, absent Doppler color flow Lobules and vessels
docrine disease MR imaging: T2-hypointense tissue Dense stromal fibro-
(thyroid), and when breast is densely fibrotic, sis with keloidal
premenopausal nonspecific stromal enhance- features and
women ment myofibroblasts
Wegener Affects persons Unilateral or Very nonspecific Vascular destructive
granuloma- known to have bilateral breast Mammography: ill-defined irregu- leukocyte infiltra-
tosis systemic disease masses lar masses tion (angiitis)
of upper and Breast abscesses US: irregular hypoechoic masses Granuloma forma-
lower respira- Necrotic lesions tion and aseptic
tory tracts and and skin ulcer- tissue necrosis
sometimes the ations
kidneys
Breast involvement
is rare
Breast sar- Affects persons Palpable mass Mammography: irregular, ill- Noncaseating
coidosis known to have Usually less inflam- defined, spiculated, or well- granulomas with
systemic disease mation circumscribed round masses or without giant
and women in Abscess formation US: irregular hypoechoic masses cells
the 3rd or 4th is uncommon
decade of life
Foreign body Affects persons Focal or diffuse Silicone: round or oval dense Foreign material is
granulomas with a his- lumps masses with rim calcifications usually obvious,
caused by tory of direct Induration on mammograms, “snowstorm” with granuloma-
silicone, breast cosmetic Breast deformity appearance on US images tous reaction and
paraffin, enhancement Pain and tender- Paraffin: irregular or round hy- variable fibrosis
or PAAG and transexual ness poechoic masses, parenchymal
injections* males Skin ulceration distortion, dystrophic or ringlike
Draining sinuses calcifications on mammograms;
Axillary lymphade- posterior shadowing mass on
nopathy if mate- US images
rial migrates PAAG: discrete fluid collections
that are denser than adjacent
tissue on mammograms; cir-
cumscribed fluid, anechoic to
hyperechoic collections with a
thick capsule, and/or patchy ar-
eas of mixed or granular echoes
on US images
*PAAG = polyacrylamide hydrogel.
346  March-April 2018 radiographics.rsna.org

Figure 13.  Classic IBC in a 62-year-old woman who presented with a large palpable mass in
the upper region of the right breast, with accompanying ipsilateral diffuse skin edema, tender-
ness, and asymmetric breast heaviness of 2 weeks’ duration. (a, b) Right-breast MLO (a) and
left-breast MLO (b) mammographic views show asymmetrically increased breast density (thin
arrow in a), extensive skin thickening (thick arrow in a), and morphologically abnormal axillary
lymph nodes (* in a) in the right breast, without a discrete breast mass or focal asymmetry.
(c) Targeted US image at the 12-o’clock position in the right breast shows an irregular, het-
erogeneously hypoechoic parallel mass (arrowhead) with overlying skin edema. FN Rad = from
nipple radial. This mass was classified as a BI-RADS category 4 lesion. Histologic analysis of US-
guided core-needle biopsy specimen revealed poorly differentiated invasive ductal carcinoma
with lymphovascular invasion, consistent with IBC.

Figure 14.  Infective mastitis with abscess in a 29-year-old woman who

presented with a tender left-breast mass with associated fevers and night
sweats. (a, b) Initial CC (a) and MLO (b) diagnostic mammographic
views show an 8.7-cm circumscribed high-density left-breast mass (ar-
row). No lymphadenopathy or skin changes are seen. Physical exami-
nation revealed a large palpable mobile and tender retroareolar mass
without overlying skin changes. (c) Targeted US image of the left breast
shows a large complex retroareolar anechoic structure with hypoechoic
fluctuant debris (arrowhead). FN = from nipple, Rad = radial. There was
mild cortical thickening of an ipsilateral axillary lymph node (not shown).
FNA biopsy yielded abundant purulent material, with the culture posi-
tive for Staphylococcus aureus. The patient continued taking the initially
prescribed antibiotics until the findings resolved. Short-term US follow-up
revealed complete resolution of the clinical and imaging findings.

of failed antibiotic therapy and the rare finding of

hydatid cysts associated with Echinococcus infec-
tion, there are no substantial clinical or imaging
characteristics that can be used to distinguish
infective mastitis from IGM.

Tuberculous Mastitis
Tuberculous mastitis is an important consid-
eration to exclude in high-risk populations and
endemic areas, as it can have marked systemic
manifestations, may constitute a public health haz-
ard, and is a contraindication for steroid therapy
(59). In addition, targeted long-course antibiotic
therapy is required. In a study in which the find-
ings in patients with IGM were compared with
those in patients who had tuberculous mastitis,
Seo et al (59) found that the most common mani-
RG  •  Volume 38  Number 2 Pluguez-Turull et al  347

Figure 15.  Mammary duct ectasia in a 60-year-old woman who presented for additional evaluation
of multiple small bilateral subareaolar breast masses. The masses were palpable, nontender, and slowly
enlarging. (a, b) CC (a) and MLO (b) diagnostic mammographic views of the right breast show small
irregular masses (arrow in a) in the right upper outer periareolar region and left upper periareolar region
of the breast. Branching tubular lucencies (arrows in b) are noted in the subareolar region of the right
breast and are better seen on the MLO view. (c) Targeted US image of the right breast shows ectatic sub-
areaolar breast ducts with dependent debris and multiple hypoechoic intraductal lesions or filling defects
(*). A classification of BI-RADS category 4 lesions was recommended so that the filling defects could be
sampled; however, at the time of US-guided core-needle biopsy, the target lesions were found to contain
swirling debris instead of filling defects, and the biopsy was not performed. The classic secretory calcifica-
tions were not identified; these are well visualized in Figure 10.

festation in both groups was a breast mass. How-
ever, they found statistically significant clinical
differences between the two patient groups: older
age at presentation in the tuberculous mastitis
group (40 vs 33.5 years, P =.018), more frequent
axillary lymphadenopathy in the tuberculous mas-
titis group (50.0% vs 20.6%, P =.048), and more
frequent mastalgia in the IGM group (84.4% vs
50.0%, P =.013). They also observed that 50%
of the patients with tuberculous mastitis had a
history of pulmonary tuberculosis, whereas the
patients in the IGM group had no such history.
Chest radiographic findings and medical history
of tuberculosis are useful for differentiating IGM
and tuberculous mastitis. Other imaging studies are
less useful, with US images showing similar com-
mon findings, including an irregular hypoechoic
mass with surrounding inflammatory changes (59).
Histopathologic analysis of breast tuberculosis
reveals caseating granulomas and positive acid-
fast stain results in cases of tuberculous mastitis,
compared with sterile noncaseating granulomas in
cases of IGM. Some authors have suggested that
acid-fast bacilli that are not detected at routine
staining analyses (without polymerase chain reac-
tion analysis) and a similar cytomorphic pattern
contribute to the underdiagnosis of tuberculous
mastitis, particularly when relevant clinical infor-
mation is not taken into account (11,33,60).
Mammary Duct Ectasia
Mammary duct ectasia is an inflammatory breast
condition that should be considered in the differ-
ential diagnosis of IGM. Mammary duct ectasia is
characterized by chronic inflammation and fibrosis
of the breast, which result in dilatation of the mam-
mary ducts (Fig 15). In contrast to premenopausal
and parous women, who most commonly develop
IGM, perimenopausal and postmenopausal women
are most often affected with mammary duct ectasia
(6,56). Mammary duct ectasia can manifest clini-
cally as a breast mass, which is typically subareolar,
with or without pain, and with unilateral or bilateral
involvement (6,56). In comparison, IGM usually
involves a tender mass and is typically peripheral
and unilateral. Other common clinical findings
include a nonbloody nipple discharge and nipple
retraction (61), which are uncommon with IGM.
Mammary duct ectasia may be more confidently
diagnosed when similar imaging features are seen
bilaterally. At mammography, the most common
finding of mammary duct ectasia is tubular or
branching retroareolar lucencies or opacities with
thick rodlike (secretory) calcifications converging
toward the nipple. Calcifications are a rare find-
ing with IGM. At US, features of mammary duct
ectasia include dilated and fluid-filled subareolar
ducts, anechoic fluid collections with debris, and
sometimes an associated intraductal mass.
348  March-April 2018 radiographics.rsna.org

Figure 16.  Diabetic fibrous mastopathy in a 45-year-old woman who presented with a nontender pal-
pable right-breast mass, which had not changed for 3 years. The mass was previously evaluated mammo-
graphically, and the patient was told that it was benign. She also had a history of end-stage renal disease
related to uncontrolled long-standing type 2 diabetes mellitus. (a, b) CC (a) and MLO (b) screening
mammographic views show a heterogeneously dense right breast with a 12-mm focal asymmetry (ar-
row) in the 10-o’clock position, 6 cm from the nipple at middle depth. (c) The patient was recalled for
diagnostic supplemental mammography, including the acquisition of an MLO compression view of the
right breast, which shows a persistent 1.9-cm irregular mass (arrow) in the 9-o’clock position. Physical
examination revealed a 2-cm firm oval mobile mass. (d) Subsequently obtained targeted US image of
the outer region of the right breast shows a 1.9-cm parallel irregular avascular mass with angulated mar-
gins and strong posterior acoustic shadowing. The mass was classified as a BI-RADS category 4 lesion.
US-guided core-needle biopsy with histologic analysis revealed stromal fibrosis with abundant collagen
deposition, consistent with our diagnosis of diabetic fibrous mastopathy.

A helpful diagnostic feature is movement of

the intraductal secretions during real-time US;
this is not commonly seen with IGM. With mam-
mary duct ectasia, branching T2-hyperintense tu-
bular structures that converge toward the nipple,
often with T1 hyperintensity due to intraductal
proteinaceous material or blood, can be seen on
MR images (62).

Diabetic Fibrous Mastopathy

If the patient has a long-standing history of
insulin-dependent diabetes or thyroid disease,
diabetic mastopathy is another condition that
should be considered in the differential diagno-
sis, as it may manifest similarly to IGM (Fig 16).
Diabetic mastopathy is a fibrous inflammatory
proliferation of the breast tissue that is character-
istically seen in premenopausal women about 20
years after the onset of diabetes. It usually mani-
fests clinically as multiple painless hard masses,
which are frequently bilateral (58,63). Similar to
the imaging findings of IGM, the imaging find-
ings of diabetic mastopathy are inconclusive and
often indistinguishable from those of malignancy.
At mammography, diabetic mastopathy is usually
seen as ill-defined masses or focal asymmetries.
US findings include multiple irregular hypoechoic
masses with absent Doppler color flow and strong
posterior acoustic shadowing, the latter of which is
related to the fibrotic component (58,63). MR im-
ages show an irregular T2-hypointense mass with
nonspecific stromal enhancement (58,64).
Wegener Granulomatosis
Wegener granulomatosis is a necrotizing granulo-
matous vasculitis that characteristically affects the
upper and lower respiratory tracts and kidneys.
Breast involvement is very unusual, with approxi-
mately 28 cases reported—mainly in women with
systemic manifestations of the disease. Primary
breast involvement is very rare. Usual clinical
manifestations include unilateral or bilateral breast
masses, breast abscesses, necrotic lesions, and
ulcerations (63). Nonspecific imaging findings
may include ill-defined irregular masses at mam-
mography and irregular hypoechoic masses at US
(58,64). A definite diagnosis of Wegener granulo-
RG  •  Volume 38  Number 2 Pluguez-Turull et al  349
Figure 17.  Bilateral silicone granulomas that manifested as
mild diffuse bilateral breast pain, fevers that began 1 month
earlier, multiple palpable mildly tender breast masses, and
bilateral symmetric inner lower quadrant skin erythema in a
27-year-old woman who was injected with breast-enhancing
material 3 years earlier. (a) MLO diagnostic mammogram of
the left breast shows multiple diffuse bilateral high-density
breast masses (*) that are small, round or oval, circumscribed,
and obscured. (b) US image of the left breast at the 3-o’clock
position, at a site of palpable abnormality, shows a parallel hy-
poechoic and avascular lesion with indistinct margins (arrow).
The snowstorm appearance of free silicone is a characteristic
heterogeneous echogenic appearance with dispersion of the
ultrasound beam.
matosis is made at histologic analysis, which reveals
vascular destructive leukocyte infiltration, granu-
loma formation, and aseptic tissue necrosis (65).
Sarcoidosis
Sarcoidosis is an immunologic disease that can
affect any organ system, but it more frequently
involves the lungs, lymph nodes, skin, eyes,
liver, and spleen. Breast sarcoidosis, defined as
breast granulomas in patients with sarcoidosis,
is very rare, accounting for less than 1% of cases
of breast disease associated with sarcoidosis. It
generally occurs in patients with well-known
systemic involvement (58,63). The most frequent
manifestation is a palpable mass in women in
the 3rd or 4th decade of life (58). In contrast to
IGM, sarcoid granulomas tend to exhibit less
inflammation and are not associated with ab-
scess formation (56). At mammography, breast
sarcoidosis has been described as irregular,
ill-defined, or spiculated masses that are often
suspicious for breast cancer. Small well-defined
round masses also have been described, and in
some cases, these may represent intramammary
lymph node involvement. Calcifications are usu-
ally absent. The most common US findings are
irregular hypoechoic masses (58,63).
Silicone, Paraffin, or PAAG Injection
Granulomas
Silicon, paraffin, or PAAG injection granulo-
mas also have a clinical manifestation similar
to that of IGM (Fig 17). Because the use of
these injectable breast materials for cosmetic
purposes has been widely proven to be unsafe,
these mimics of IGM have become less common.
A clinical history of direct cosmetic enhance-
ment is often the strongest clue to the diagnosis
of PAAG injection–related granulomas. In the
case of liquid silicone injections, fibrous silicone
masses commonly appear at mammography as
well-defined, round, peripherally calcified lesions,
which sometimes develop spiculated margins
owing to developing fibrosis. This appearance is
usually suspicious for breast cancer. Extracapsu-
lar silicone from ruptured breast implants may
appear with silicone granulomas (or siliconomas)
(Fig 17), which can be found in the breast and
regional breast lymph nodes. On US images,
silicone granulomas have a typical snowstorm ap-
pearance of diffuse highly echogenic nodules with
a well-defined anterior margin and characteristic
posterior shadowing (66).
Injected paraffin appears on mammograms as
circumscribed noncalcified masses that elicit sinus
tract formation and tissue necrosis. When paraffin
granulomas, or paraffinomas, develop, common
mammographic findings are irregular masses,
parenchymal distortion, streaky opacities, and
dystrophic or ringlike calcifications. On US im-
ages, paraffinomas appear as irregular hypoechoic
masses with posterior acoustic shadowing. They
are sometimes associated with architectural distor-
tion, reflecting fibrosis. PAAG collections are dif-
ficult to identify mammographically because they
are isodense to breast tissue and remain incon-
spicuous in asymptomatic patients. On US images,
PAAG deposits appear as circumscribed anechoic
to hypoechoic fluid collections with granular and
patchy internal echoes (67). In cases of symp-
tomatic PAAG injections, the collections appear
to be denser than the adjacent breast tissue, with
a thick surrounding capsule. The MR imaging
appearances of these lesions are beyond the scope
of this review article (68). Although direct trauma
is another cause of mastitis, it usually involves no
clinical challenges (69).
Histopathologic Analysis of IGM
A distinctive histopathologic feature of IGM
is noncaseating lobulocentric granulomatous
350  March-April 2018 radiographics.rsna.org

Figure 18.  Classic histopathologic findings of IGM in multiple lesions. (a) Photomicrograph shows the early changes
seen with IGM, which consist of lobulocentric lymphocytic inflammatory activity with destruction of central acini
(arrows) and relative preservation of the peripheral acini. (Hematoxylin-eosin stain; original magnification, 3200.)
(b) Photomicrograph shows the edge of a lobule, with lymphocytic and granulomatous inflammation. A lymphocyte-
permeated damaged acinus (white arrowhead) and Langerhans giant cell (black arrowhead) are seen, and surrounding
fibrosis (arrow) is forming. (Hematoxylin-eosin stain; original magnification, 3400.)

inflammation. This inflammation is usually

confined to—or at least centered in—the breast
lobules (acini), generally sparing major ducts and
surrounding fat tissue. Therefore, IGM is also
frequently referred to as granulomatous lobu-
lar mastitis so that it can be distinguished from
granulomatous periductal mastitis (2). Within
the inflamed lobules there may be lymphocytes,
plasma cells, eosinophils, and (occasionally)
neutrophilic microabscesses. Sterile granulomas
also are seen invariably and are composed of
epithelioid histiocytes and Langerhans giant cells,
which may also have a suppurative center (Fig
18) (1,2,70). The gross pathologic appearance
of IGM is nonspecific and depends on the lesion
size and the varying degrees of inflammation, Figure 19.  Gross pathologic appearance of IGM. Photograph
of cut gross specimen excised at surgery from the patient in
fibrosis, and possible abscess formation (Fig 19). Figure 11 shows a central fibrous breast lesion (arrow) with
At routine initial histopathologic assessment ill-defined borders.
of an IGM specimen obtained at core-needle
biopsy, one should recognize an inflammatory
process. The subsequent examination should be (DCIS) (72). DCIS is characterized by ductal
focused primarily on determining the underlying cells with a clonal appearance or marked cyto-
cause, excluding infection, and excluding under- logic atypia lining the calcified ducts.
lying malignancy. At histologic evaluation, one Although IGM may have features that are
will find that IBC, despite the name, is actually clinically suspicious for malignancy, the histo-
not an inflammatory process; it is character- logic analysis–based differential diagnosis consists
ized by an extensive angiolymphatic spread of mainly of other inflammatory processes—infec-
carcinoma that produces a pseudoinflammatory tion in particular. When granulomatous inflam-
clinical and imaging appearance (71). As with mation is present, microbial stain analyses usually
any breast biopsy, the biopsy procedure involves are performed. Gram stains may be performed
a search for invasive breast carcinoma, which is when acute inflammation, with or without ab-
characterized by infiltrating irregular and angu- scess formation, is observed. Infection caused
lated nests of epithelial cells with cytologic atypia by mycobacteria and fungal agents can lead to
within a desmoplastic background. It also in- granulomatous mastitis; however, in this setting,
cludes a search for microcalcifications, a finding granulomas usually demonstrate central case-
that may be seen with inflammation, fat necrosis, ation, a finding that is absent with IGM (73–75).
fibrocystic changes, and ductal carcinoma in situ Special stains that demonstrate acid-fast bacilli
RG  •  Volume 38  Number 2 Pluguez-Turull et al  351
Figure 20.  Algorithm for IGM management at University of
Texas Health at San Antonio. The flowchart outlines the proto-
col for treating patients with antibiotic-resistant breast inflam-
matory changes. If the original manifesting clinical feature is a
suspicious palpable finding or the primary physician did not ex-
amine the patient previously, then initial breast imaging evalu-
ation results may indicate the need for antibiotic treatment
with short-interval follow-up. Evaluation usually includes lim-
ited physical examination, US, and possibly subsequent mam-
mographically guided interventions to drain fluid collections
and examine core-needle biopsy samples from lesions with
suspicious imaging features. In most cases, imaging-guided bi-
opsy and histopathologic analysis will enable differentiation of
IGM, malignancy, atypical or resistant infection, foreign body
reactions, and other inflammatory breast diseases. Biopsy-
proven IGM with mild symptoms or that is encountered inci-
dentally can be surveyed with US every 3–6 months and with
annual mammography, as clinically indicated according to the
patient’s age and imaging findings. In the remaining patients,
steroid therapy can be instituted, with follow-up imaging fo-
cused on treatment response. Unsuccessful or contraindicated
steroid therapy should prompt second-line medication man-
agement involving the use of alternative immunosuppressive
(methotrexate) or prolactin-lowering (bromocriptine) agents.
If medication therapy is unsuccessful or side effects preclude
the use of conservative treatment with medication, then surgi-
cal management is offered.
(acid-fast and Fite stains) and fungi (periodic
acid–Schiff and Gomori methenamine-silver
stains) can be of further assistance in this distinc-
tion and are used routinely whenever granulomas
are seen. Bacterial infection may cause an abscess
characterized by tissue necrosis and suppura-
tive inflammation, without lobulocentricity or
granulomas. Alternatively, these infections can
cause only neutrophilic inflammation without an
abscess; this is diagnosed simply as acute mastitis.
In this setting, Gram stains are rarely performed
in deference to microbiologic studies (76,77).
Noninfectious inflammatory processes in-
cluded in the histopathologic differential diag-
noses of IGM include ductal ectasia, plasma cell
mastitis, foreign body granulomas, and sarcoid-
osis. The histologic finding of ductal ectasia is
cystic duct dilatation with surrounding lymphoid
and histiocytic inflammation that is distinctly
ductocentric, rather than lobulocentric, and
nongranulomatous (78,79). With plasma cell
mastitis, a granulomatous component is fre-
quently present, but the inflammation is equally
prominent in periductal, perilobular, and stromal
components of the breast tissue and is heavily
plasmacytic (80). Foreign body granulomas may
be seen in association with a variety of processes,
including fat necrosis and injected foreign breast-
augmenting material. However, in such instances
the agent eliciting the reaction typically is obvi-
ous and the inflammation is stromal (81). Last,
sarcoidosis may be difficult to distinguish from
IGM with histologic analysis alone; however, the
granulomas are as likely to be in the interlobular
stroma as they are to be in the lobule. In addi-
tion, because isolated mammary sarcoidosis is
infrequently reported, extramammary findings of
sarcoidosis are present (82,83).
Management of IGM
The treatment of IGM remains controversial.
Because randomized clinical trials to determine
the optimal treatment of IGM are lacking, man-
agement is generally tailored to each patient’s
clinical manifestations. A representative algo-
rithm for the diagnosis and management of IGM
at University of Texas Health at San Antonio is
presented in Figure 20 (11). Treatment options
include conservative measures such as close
regular surveillance (3) and medication therapy
with corticosteroids (84) or methotrexate (85),
and the more aggressive approach of wide local
surgical excision (86). These approaches to man-
aging IGM are briefly described in the following
sections.
Surveillance
In a retrospective study of patients with histo-
pathologically confirmed IGM, who were under-
going conservative treatment involving regular
follow-up every 2–3 months, Lai et al (3) found
complete regression of IGM in 50% of the pa-
tients after a period of 2–24 months (mean, 14.5
months). The disease in the remaining 50% of
the patients, which was followed up with clinical
examination and imaging, remained static, lead-
ing the authors to recommend surveillance. The
studies in which surveillance has been advocated
352  March-April 2018 radiographics.rsna.org
have involved patients who presented predomi-
nantly with reports of a palpable mass without
other discomforting symptoms that are character-
istic of IGM. Investigators in additional studies
(4,6) have reported a spontaneous burnout of
IGM lesions in symptomatic patients after 6–12
months, irrespective of the treatment adminis-
tered. Close surveillance alone may be appropri-
ate for a subgroup of patients, especially those
in whom IGM is discovered incidentally during
screening mammography and those in whom
IGM manifests as a painless or mildly tender
palpable mass. (3) There is currently no consen-
sus regarding the radiologic surveillance of IGM;
however, some authors have endorsed a procotol
involving mammography performed annually and
US performed every 3–6 months after the acute
episode until the disease resolves (3,17).
Medication Therapy
Given the far more common incidence of infec-
tive mastitis and the overlapping imaging findings
of IGM and infective mastitis, most patients who
are examined owing to suspicion for mastitis are
treated with empirical antibiotic therapy targeting
common pathogens such as Staphylococcus (in-
cluding methicillin-resistant variants) and Strep-
tococcus. If necessary, further antibiotic therapy is
often tailored according to the clinical response.
A partial response or no response, which may
suggest the presence of a resistant or atypical
infection (eg, with fungi or C kroppenstedtii), is an
indication for prolonged therapy (16). For this
initial assessment performed by the primary care
provider, breast imaging or an interventional pro-
cedure usually is not required, unless a concomi-
tant abscess is suspected or the antibiotic therapy
was unsuccessful.
IGM itself is generally a sterile condition, and
there are insufficient data to support routine use
of antibiotics. In fact, the diagnosis of IGM is
often reached after failed prolonged antibiotic
therapy for presumed infectious mastitis. The use
of antibiotics is limited to patients who have a
bacterial infection in addition to IGM (5,87). In
a retrospective review of patients with histologi-
cally confirmed IGM who presented to a large
tertiary care center during a 2-year period, none
of the patients responded to multiple courses of
antibiotics (87). Even in patients with concomi-
tant infections or abscesses, the use of antibiot-
ics alone is often unsuccessful in controlling the
disease or improving symptoms (5).
Corticosteroids have been shown to be an
effective first-line therapy for patients with
histopathologically proven symptomatic IGM
(5,88). Aghajanzadeh et al (5) performed the
largest study of IGM to date, which involved a
retrospective analysis of the data on 206 patients
with histologically confirmed IGM in northern
Iran. All of the patients were initially treated with
antibiotics (cloxacillin, cephalexin, ciprofloxacin,
or clindamycin); this resulted in resolution of
symptoms in only six (3%) patients. Seventy-two
percent of the 200 patients in whom antibiotic
therapy failed responded to 10–20 mg of prednis-
olone taken three times daily, with the maximal
treatment duration being 6 months. Similarly, in
a meta-analysis of all studies of IGM published
between 1972 and 2010 (88), the investigators
reported a full recovery in 72% of the patients
treated with corticosteroids. Studies on maximal
treatment duration have endorsed a full recovery
within an average of 4–10 months (89).
A clear disadvantage of corticosteroid therapy
is the long-term use required to achieve complete
resolution of the masses, which can take months
to years after initiation of therapy (89). These
long time lines not only often overlap with the
described self-burnout rates of IGM lesions (3,6),
but they also predispose the patient to known side
effects, including Cushing syndrome, osteopenia,
hyperglycemia, weight gain, and dyspepsia (5). A
consensus regarding the duration, dose, and route
of administration of corticosteroids for treatment
of IGM remains to be established (90). Aghajanza-
deh et al (5) proposed an IGM treatment strategy
of 30–60 mg of oral prednisolone daily for 4 weeks
with decreasing doses over periods of 3, 5, and 6
weeks. When this protocol was unsuccessful, the
addition of methotrexate at a dose of 7.5–10.0 mg
per week was recommended. If this strategy also
is unsuccessful, prolactin-lowering medication
such as bromocriptine (5–10 mg/day), a dopamine
agonist, together with glucocorticoids may be con-
sidered for a combination therapy. Bromocriptine
has been shown to be most effective in cases of
refractory IGM with concomitant hyperprolactin-
emic states (9).
Methotrexate is an immunosuppressant that
has been found to be effective, with or without si-
multaneous corticosteroid therapy, for treatment
of IGM (85). It is particularly useful in cases
of steroid-resistant IGM and in patients who
develop steroid-associated glucose intolerance or
Cushing syndrome (88). Azathioprine is an im-
munosuppressant that can be used alternatively
in patients who develop methotrexate-induced
pneumonitis (88). For use of bromocriptine,
methotrexate, and prolonged oral steroid therapy,
collaboration with endocrinology and/or rheuma-
tology physicians is usually required.
Surgical Management
Owing to frequent difficulty in distinguishing
IGM from breast cancer, some institutions prefer
RG  •  Volume 38  Number 2 Pluguez-Turull et al  353
surgery as the first-line treatment, regardless
of the FNA and/or core-needle biopsy results
(91,92). Use of a surgical approach may lead to
repeated surgical procedures, increasing the risk
of multiple scars. It may also cause nipple and
breast distortion without symptom resolution or
lead to mastectomy (93). Therefore, surgery is
generally reserved for cases of refractory and/or
recurrent IGM. Surgical management can range
from abscess drainage to wide-margin excision of
an IGM inflammatory mass, and in extreme cases
of widespread disease, to mastectomy (94). Most
commonly, the surgical approach involves wide
local excision of the inflammatory mass and the
associated pathologic ductal system. In a study to
assess this approach (95), 93% of patients dem-
onstrated a rapid recovery during an 18-month
period, without recurrence.
For patients with larger inflammatory masses
encompassing 20%–50% of the breast in whom
medication treatment was unsuccessful, thera-
peutic mammoplasty has been proposed as an
option. In a study conducted by Ahmed and Abd
El Maksoud (86), therapeutic mammoplasty was
performed in 13 patients with large histologically
confirmed IGM masses, with the option to per-
form a contralateral reduction to achieve symme-
try and optimal aesthetics. The authors reported
recurrent IGM in two (15%) of the 13 patients
at 16 and 24 months. An assessment of patient
satisfaction revealed that 10 (77%) patients were
satisfied with the results of the surgery.
In contrast, investigators in other studies (96)
have reported recurrence rates as high as 73%
after wide surgical excision of IGM masses, with
multiple procedures required to achieve complete
remission. The discrepancy in recurrence rates af-
ter surgery reported in the literature is large, and
it is difficult to discern whether it is related to the
surgical techniques used or other characteristics
of the disease. The lesion size at diagnosis has
been proposed as an indicator of resistance to
medication and surgical therapy (97). Some plas-
tic surgeons suggest a delay between the initial
wide excision of the inflammatory mass and the
cosmetic reconstruction to account for the prob-
ability of recurrence (96).
Medication Therapy versus
Surgical Treatment
Prospective clinical trials involving direct com-
parisons of medication and surgical treatments of
IGM are scarce. In a study in which these treat-
ment strategies were compared, Salehi et al (98)
found pharmaceutical therapy with azithromycin
and prednisolone to be effective for treatment
and prevention of IGM relapse for a period of
12 months. In that study, relapse was found to
be primarily dependent on the number of lesions
present at the time of diagnosis, regardless of the
treatment method. However, in the largest study
to date involving a comparson of medication
versus surgical management of IGM, Yabanoğlu
et al (99) found the recurrence rate to be much
higher among the patients who were treated
conservatively. They also reported recovery rates
to be much faster (within 1–5 months) after wide
local excision (99). In comparison, the patients
who underwent medication therapy needed 1–15
months to recover.
Conclusion
IGM is a benign chronic inflammatory breast
disease characterized by noncaseating granulo-
matous inflammation, and it often occurs in par-
ous women of childbearing age. It is frequently
associated with lactation or hyperprolactinemia,
and an increased incidence in nonwhite persons
has been proposed (6,17,23,29). The most com-
mon clinical manifestation of IGM is a unilateral
palpable tender breast mass (5,10). It often mani-
fests with abscess formation and cutaneous sinus-
draining sterile inflammatory components.
The imaging findings of IGM are nonspe-
cific and overlap with those of other benign and
malignant breast entities. The most common
mammographic finding is focal asymmetry, but
manifestation as an irregular mass or asymmetri-
cally increased breast density (global asymme-
try) also has been described (11,13,17,25,32,
39,41,45–47). An irregular hypoechoic parallel
mass with tubular extensions is a common US
finding; however, a variety of US appearances
have been described (5,11,13,17,25,26,32,39,44,
45,48). In the few published studies of IGM at
MR imaging, a sensitivity of 100% is reported,
with a heterogeneously or rim-enhancing irregu-
lar mass with type I kinetic properties accompa-
nied by segmental or regional NME described as
common findings (13,17,25,32,39,41).
IGM has clinical and imaging characteristics
that frequently overlap with those of IBC and
other inflammatory breast diseases. A suggestive
clinical manifestation, palpable axillary lymph-
adenopathy, and unilateral breast enlargement
with extensive skin changes suggest IBC (52). A
failed response to antibiotics suggests that the
disease is not a simple infective mastitis; how-
ever, biopsy with microbial staining and culture
analysis is required to distinguish atypical or
resistant infective mastitis with or without abscess.
Tuberculous mastitis has a histologic appearance
similar to that of IGM, and the diagnosis of this
entity is further complicated by the somewhat
limited sensitivity of acid-fast staining analyses.
However, it may be differentiated on the basis of
354  March-April 2018 radiographics.rsna.org
a history of tuberculosis and the findings of ad-
ditional mycobacterial polymerase chain reaction
analysis, as endorsed by some authors (59). Mam-
mary duct ectasia is more often bilateral, subareo-
lar, and less symptomatic compared with IGM;
imaging features include dilated retroareolar
ducts with characteristic histologic features (6).
Diabetic fibrous mastopathy lesions tend to be
painless, are often bilateral, and occur in patients
with a history of long-standing insulin-dependent
diabetes or thyroid disease, with characteristic
posterior acoustic shadowing at US (63).
Additional, rarer entities include Wegener
granulomatosis, which has similar imaging
findings but is more likely to be associated with
necrotic skin ulcerations and fluid collections,
with characteristic necrotizing vascular leuko-
cyte infiltration seen at histologic analysis (63).
Sarcoidosis may have clinical, imaging, and
histologic findings similar to those of IGM, and,
thus, clinicopathologic analysis is required for the
diagnosis. Injected foreign body material lesions
are differentiated on the basis of the medical
history and the finding of foreign body material
at histologic analysis (81). Given the nonspecific
clinical manifestations and imaging findings
of IGM, the diagnosis is usually based on the
presence of specific histopathologic findings in
core-needle biopsy specimens and the exclusion
of other causes of inflammatory breast disease.
The definitive treatment of IGM remains
controversial, and options include surveillance,
medication therapy, and surgery. Without medi-
cation or surgical treatment, IGM has a persis-
tent or natural recurrent course, with at least
half of affected patients demonstrating eventual
lesion burnout at 1–2 years after presentation
(3,4,6). Therefore, imaging surveillance has been
endorsed for cases of incidentally encountered or
mildly symptomatic IGM. Although there is no
consensus regarding the radiologic surveillance
of IGM, examination of the affected breast with
mammography annually and with US every 3–6
months after the acute episode until the disease
has resolved has been suggested for patients who
opt for expectant management (3,17). In most
other cases, oral steroid therapy is instituted as
the first-line therapy, with complete recovery
rates of up to 72% and maximal treatment dura-
tions of 6–10 months reported in the literature
(88). Certain patients may require combination
therapy or monotherapy with methotrexate and/
or bromocriptine. There is insufficient evidence
to support the routine use of antibiotics in cases
of biopsy-proven IGM, even in the presence of an
abscess (5,87). Surgical management with wide
local excision or mastectomy should be consid-
ered after failed conservative management in
cases of refractory or moderate to severe recur-
rent disease. However, this approach is frequently
associated with recurrence and the need for mul-
tiple procedures to achieve remission (86,96,97).
At our institution, imaging surveillance for mildly
symptomatic or incidentally discovered lesions
and conservative treatment with oral steroids,
methotrexate, and/or bromocriptine before con-
sidering surgical management are recommended.
The role of imaging in the evaluation of
biopsy-confirmed IGM is to enable the clinician
to (a) establish the multiplicity and location of
IGM lesions, (b) document the size of lesions,
(c) identify abscess formation and the associated
possibility of intervention, (d) evaluate the stabil-
ity of or interval change in lesions, (e) evaluate
treatment response, and (f) identify metachro-
nous disease and local recurrence with imaging
surveillance.
Acknowledgments.—The authors thank Meeghan Lautner,
MD, and Ismail Jatoi, MD, University of Texas Health at San
Antonio–Medical Arts and Research Center Surgery Depart-
ment, for their contributions to the management portions
of this article. In addition, the authors acknowledge Laura
Maaske for creating Figure 1 and Jonathan Sumner for his
work with the annotations and formatting the images.
References
1. Kessler E, Wolloch Y. Granulomatous mastitis: a le-
sion clinically simulating carcinoma. Am J Clin Pathol
1972;58(6):642–646.
2. Going JJ, Anderson TJ, Wilkinson S, Chetty U. Granuloma-
tous lobular mastitis. J Clin Pathol 1987;40(5):535–540.
3. Lai EC, Chan WC, Ma TK, Tang AP, Poon CS, Leong HT.
The role of conservative treatment in idiopathic granuloma-
tous mastitis. Breast J 2005;11(6):454–456.
4. Yukawa M, Watatani M, Isono S, et al. Management of
granulomatous mastitis: a series of 13 patients who were
evaluated for treatment without corticosteroids. Int Surg
2015;100(5):774–782.
5. Aghajanzadeh M, Hassanzadeh R, Alizadeh Sefat S, et al.
Granulomatous mastitis: presentations, diagnosis, treatment
and outcome in 206 patients from the north of Iran. Breast
2015;24(4):456–460.
6. Al-Khaffaf B, Knox F, Bundred NJ. Idiopathic granulo-
matous mastitis: a 25-year experience. J Am Coll Surg
2008;206(2):269–273.
7. Altintoprak F, Karakece E, Kivilcim T, et al. Idiopathic
granulomatous mastitis: an autoimmune disease? Sci World
J 2013;2013:148727.
8. Bani-Hani KE, Yaghan RJ, Matalka II, Shatnawi NJ. Idio-
pathic granulomatous mastitis: time to avoid unnecessary
mastectomies. Breast J 2004;10(4):318–322.
9. Nikolaev A, Blake CN, Carlson DL. Association between
hyperprolactinemia and granulomatous mastitis. Breast J
2016;22(2):224–231.
10. Baslaim MM, Khayat HA, Al-Amoudi SA. Idiopathic
granulomatous mastitis: a heterogeneous disease with
variable clinical presentation. World J Surg 2007;31(8):
1677–1681.
11. Hovanessian Larsen LJ, Peyvandi B, Klipfel N, Grant E, Iy-
engar G. Granulomatous lobular mastitis: imaging, diagnosis,
and treatment. AJR Am J Roentgenol 2009;193(2):574–581.
12. DeHertogh DA, Rossof AH, Harris AA, Economou SG.
Prednisone management of granulomatous mastitis. N Engl
J Med 1980;303(14):799–800.
13. Oztekin PS, Durhan G, Nercis Kosar P, Erel S, Hucumenoglu
S. Imaging findings in patients with granulomatous mastitis.
Iran J Radiol 2016;13(3):e33900.
RG  •  Volume 38  Number 2 Pluguez-Turull et al  355
14. Bland KI, Copeland EM. The breast: comprehensive man-
agement of benign and malignant diseases. Philadelphia, Pa:
Saunders/Elsevier, 2009.
15. Kim J, Tymms KE, Buckingham JM. Methotrexate in
the management of granulomatous mastitis. ANZ J Surg
2003;73(4):247–249.
16. Taylor GB, Paviour SD, Musaad S, Jones WO, Holland DJ. A
clinicopathological review of 34 cases of inflammatory breast
disease showing an association between corynebacteria infection
and granulomatous mastitis. Pathology 2003;35(2):109–119.
17. Gautier N, Lalonde L, Tran-Thanh D, et al. Chronic granu-
lomatous mastitis: imaging, pathology and management. Eur
J Radiol 2013;82(4):e165–e175.
18. Reddy KM, Meyer CE, Nakdjevani A, Shrotria S. Idio-
pathic granulomatous mastitis in the male breast. Breast J
2005;11(1):73.
19. Al Manasra AR, Al-Hurani MF. Granulomatous mastitis: a rare
cause of male breast lump. Case Rep Oncol 2016;9(2):516–519.
20. Tedeschi LG, McCarthy PE. Involutional mammary duct
ectasia and periductal mastitis in a male. Hum Pathol
1974;5(2):232–236.
21. Al-Masad JK. Mammary duct ectasia and periductal mastitis
in males. Saudi Med J 2001;22(11):1030–1033.
22. Oran EŞ, Gürdal SÖ, Yankol Y, et al. Management of idio-
pathic granulomatous mastitis diagnosed by core biopsy: a
retrospective multicenter study. Breast J 2013;19(4):411–418.
23. Mohammed S, Statz A, Lacross JS, et al. Granulomatous
mastitis: a 10 year experience from a large inner city county
hospital. J Surg Res 2013;184(1):299–303.
24. Kok KY, Telisinghe PU. Granulomatous mastitis: pre-
sentation, treatment and outcome in 43 patients. Surgeon
2010;8(4):197–201.
25. Al-Khawari HA, Al-Manfouhi HA, Madda JP, Kovacs A,
Sheikh M, Roberts O. Radiologic features of granulomatous
mastitis. Breast J 2011;17(6):645–650.
26. Yildiz S, Aralasmak A, Kadioglu H, et al. Radiologic find-
ings of idiopathic granulomatous mastitis. Med Ultrason
2015;17(1):39–44.
27. Lin CH, Hsu CW, Tsao TY, Chou J. Idiopathic granulo-
matous mastitis associated with risperidone-induced hyper-
prolactinemia. Diagn Pathol 2012;7(1):2.
28. Erhan Y, Veral A, Kara E, et al. A clinicopthologic study of
a rare clinical entity mimicking breast carcinoma: idiopathic
granulomatous mastitis. Breast 2000;9(1):52–56.
29. Skandarajah A, Marley L. Idiopathic granulomatous masti-
tis: a medical or surgical disease of the breast? ANZ J Surg
2015;85(12):979–982.
30. Pandey TS, Mackinnon JC, Bressler L, Millar A, Marcus
EE, Ganschow PS. Idiopathic granulomatous mastitis: a
prospective study of 49 women and treatment outcomes with
steroid therapy. Breast J 2014;20(3):258–266.
31. Centers for Disease Control and Prevention (CDC).
Idiopathic granulomatous mastitis in Hispanic women:
Indiana, 2006-2008. MMWR Morb Mortal Wkly Rep
2009;58(47):1317–1321.
32. Fazzio RT, Shah SS, Sandhu NP, Glazebrook KN. Idiopathic
granulomatous mastitis: imaging update and review. Insights
Imaging 2016;7(4):531–539.
33. Tse GM, Poon CS, Ramachandram K, et al. Granulomatous
mastitis: a clinicopathological review of 26 cases. Pathology
2004;36(3):254–257.
34. Martínez-Parra D, Nevado-Santos M, Meléndez-Guerrero
B, García-Solano J, Hierro-Guilmain CC, Pérez-Guillermo
M. Utility of fine-needle aspiration in the diagnosis of
granulomatous lesions of the breast. Diagn Cytopathol
1997;17(2):108–114.
35. Akahane K, Tsunoda N, Kato M, et al. Therapeutic strategy
for granulomatous lobular mastitis: a clinicopathological study
of 12 patients. Nagoya J Med Sci 2013;75(3-4):193–200.
36. Kuba S, Yamaguchi J, Ohtani H, Shimokawa I, Maeda S,
Kanematsu T. Vacuum-assisted biopsy and steroid therapy
for granulomatous lobular mastitis: report of three cases.
Surg Today 2009;39(8):695–699.
37. Lepori D. Inflammatory breast disease: the radiologist’s role.
Diagn Interv Imaging 2015;96(10):1045–1064.
38. Yeh ED, Jacene HA, Bellon JR, et al. What radiologists need
to know about diagnosis and treatment of inflammatory
breast cancer: a multidisciplinary approach. RadioGraphics
2013;33(7):2003–2017.
39. Poyraz N, Emlik GD, Batur A, Gundes E, Keskin S.
Magnetic resonance imaging features of idiopathic granu-
lomatous mastitis: a retrospective analysis. Iran J Radiol
2016;13(3):e20873.
40. Harvey JA, Mahoney MC, Newell MS, et al. ACR appro-
priateness criteria palpable breast masses. J Am Coll Radiol
2016;13(11S):e31–e42.
41. Dursun M, Yilmaz S, Yahyayev A, et al. Multimodality
imaging features of idiopathic granulomatous mastitis:
outcome of 12 years of experience. Radiol Med (Torino)
2012;117(4):529–538.
42. Sripathi S, Ayachit A, Bala A, Kadavigere R, Kumar S. Id-
iopathic granulomatous mastitis: a diagnostic dilemma for
the breast radiologist. Insights Imaging 2016;7(4):523–529.
43. Rieber A, Tomczak RJ, Mergo PJ, Wenzel V, Zeitler H,
Brambs HJ. MRI of the breast in the differential diagnosis
of mastitis versus inflammatory carcinoma and follow-up. J
Comput Assist Tomogr 1997;21(1):128–132.
44. Lee JH, Oh KK, Kim EK, Kwack KS, Jung WH, Lee HK.
Radiologic and clinical features of idiopathic granulomatous
lobular mastitis mimicking advanced breast cancer. Yonsei
Med J 2006;47(1):78–84.
45. Memis A, Bilgen I, Ustun EE, Ozdemir N, Erhan Y,
Kapkac M. Granulomatous mastitis: imaging findings
with histopathologic correlation. Clin Radiol 2002;57(11):
1001–1006.
46. Yilmaz E, Lebe B, Usal C, Balci P. Mammographic and
sonographic findings in the diagnosis of idiopathic granulo-
matous mastitis. Eur Radiol 2001;11(11):2236–2240.
47. Ozturk M, Mavili E, Kahriman G, Akcan AC, Ozturk F.
Granulomatous mastitis: radiological findings. Acta Radiol
2007;48(2):150–155.
48. Handa P, Leibman AJ, Sun D, Abadi M, Goldberg A.
Granulomatous mastitis: changing clinical and imaging
features with image-guided biopsy correlation. Eur Radiol
2014;24(10):2404–2411.
49. Dershaw DD, Moore MP, Liberman L, Deutch BM. Inflam-
matory breast carcinoma: mammographic findings. Radiology
1994;190(3):831–834.
50. Tardivon AA, Viala J, Corvellec Rudelli A, Guinebretiere JM,
Vanel D. Mammographic patterns of inflammatory breast
carcinoma: a retrospective study of 92 cases. Eur J Radiol
1997;24(2):124–130.
51. An JK, Woo JJ, Lee SA. Non-puerperal mastitis masking
pre-existing breast malignancy: importance of follow-up
imaging. Ultrasonography 2016;35(2):159–163.
52. Dawood S, Cristofanilli M. Inflammatory breast cancer:
what progress have we made? Oncology (Williston Park)
2011;25(3):264–270, 273.
53. Robertson FM, Bondy M, Yang W, et al. Inflammatory breast
cancer: the disease, the biology, the treatment. CA Cancer
J Clin 2010;60(6):351–375.
54. Wang L, Wang D, Fei X, et al. A rim-enhanced mass with
central cystic changes on MR imaging: how to distinguish
breast cancer from inflammatory breast diseases? PLoS One
2014;9(3):e90355.
55. Kamal RM, Hamed ST, Salem DS. Classification of inflam-
matory breast disorders and step by step diagnosis. Breast J
2009;15(4):367–380.
56. D’Alfonso TM, Ginter PS, Shin SJ. A review of inflamma-
tory processes of the breast with a focus on diagnosis in core
biopsy samples. J Pathol Transl Med 2015;49(4):279–287.
57. Kasales CJ, Han B, Smith JS Jr, Chetlen AL, Kaneda HJ,
Shereef S. Nonpuerperal mastitis and subareolar abscess of
the breast. AJR Am J Roentgenol 2014;202(2):W133–W139.
58. Sabaté JM, Clotet M, Gómez A, De Las Heras P, Tor-
rubia S, Salinas T. Radiologic evaluation of uncommon
inflammatory and reactive breast disorders. RadioGraphics
2005;25(2):411–424.
59. Seo HR, Na KY, Yim HE, et al. Differential diagnosis in
idiopathic granulomatous mastitis and tuberculous mastitis.
J Breast Cancer 2012;15(1):111–118.
60. Akcan A, Akyildiz H, Deneme MA, Akgun H, Aritas Y.
Granulomatous lobular mastitis: a complex diagnostic and
therapeutic problem. World J Surg 2006;30(8):1403–1409.
356  March-April 2018 radiographics.rsna.org
61. Liu L, Zhou F, Wang P, et al. Periductal mastitis: an inflam-
matory disease related to bacterial infection and consequent
immune responses? Mediators Inflamm 2017;2017:5309081.
62. Da Costa D, Taddese A, Cure ML, Gerson D, Poppiti R,
Esserman LE. Common and unusual diseases of the nipple-
areolar complex. RadioGraphics 2007;27(suppl 1):S65–S77.
63. Dilaveri CA, Mac Bride MB, Sandhu NP, Neal L, Ghosh
K, Wahner-Roedler DL. Breast manifestations of systemic
diseases. Int J Womens Health 2012;4:35–43.
64. Wong KT, Tse GM, Yang WT. Ultrasound and MR imaging
of diabetic mastopathy. Clin Radiol 2002;57(8):730–735.
65. Pambakian H, Tighe JR. Breast involvement in Wegener’s
granulomatosis. J Clin Pathol 1971;24(4):343–347.
66. Harris KM, Ganott MA, Shestak KC, Losken HW, Tobon
H. Silicone implant rupture: detection with US. Radiology
1993;187(3):761–768.
67. Yang N, Muradali D. The augmented breast: a pictorial
review of the abnormal and unusual. AJR Am J Roentgenol
2011;196(4):W451–W460.
68. Khedher NB, David J, Trop I, Drouin S, Peloquin L, Lalonde
L. Imaging findings of breast augmentation with injected
hydrophilic polyacrylamide gel: patient reports and literature
review. Eur J Radiol 2011;78(1):104–111.
69. Greydanus DE, Omar H, Pratt HD. The adolescent female
athlete: current concepts and conundrums. Pediatr Clin
North Am 2010;57(3):697–718.
70. Fletcher A, Magrath IM, Riddell RH, Talbot IC. Granu-
lomatous mastitis: a report of seven cases. J Clin Pathol
1982;35(9):941–945.
71. Amparo RS, Angel CD, Ana LH, et al. Inflammatory breast
carcinoma: pathological or clinical entity? Breast Cancer Res
Treat 2000;64(3):269–273.
72. Gümüş H, Mills P, Fish D, et al. Breast microcalcification:
diagnostic value of calcified and non-calcified cores on speci-
men radiographs. Breast J 2013;19(2):156–161.
73. Pereira FA, Mudgil AV, Macias ES, Karsif K. Idio-
pathic granulomatous lobular mastitis. Int J Dermatol
2012;51(2):142–151.
74. Binesh F, Kargar S, Zahir ST, Behniafard N, Navabi H,
Arefanian S. Idiopathic granulomatous mastitis: a clinico-
pathological review of 22 cases. J Clin Exp Pathol 2014;4:157.
75. Zhou F, Yu LX, Ma ZB, Yu ZG. Granulomatous lobular
mastitis. Chronic Dis Transl Med 2016;2(1):17–21.
76. Casas CM, Pérez M, Alados JC, et al. Nonpuerperal breast
infection. Infect Dis Obstet Gynecol 1995;3(2):64–66.
77. Petrek J. Postmenopausal breast abscess. South Med J
1982;75(10):1198–1200.
78. Dixon JM, Anderson TJ, Lumsden AB, Elton RA, Rob-
erts MM, Forrest AP. Mammary duct ectasia. Br J Surg
1983;70(10):601–603.
79. Webb AJ. Mammary duct ectasia: periductal mastitis complex.
Br J Surg 1995;82(10):1300–1302.
80. Adair F. Plasma cell mastitis: a lesion simulating mammary
carcinoma—a clinical and pathologic study with a report of
10 cases. Arch Surg 1933;26(5):735–749.
81. Tan PH, Lai LM, Carrington EV, et al. Fat necrosis of the
breast: a review. Breast 2006;15(3):313–318.
82. Banik S, Bishop PW, Ormerod LP, O’Brien TE. Sarcoidosis
of the breast. J Clin Pathol 1986;39(4):446–448.
This journal-based SA-CME activity has been approved for AMA PRA Category 1 Credit . See www.rsna.org/education/search/RG.
83. Fitzgibbons PL, Smiley DF, Kern WH. Sarcoidosis present-
ing initially as breast mass: report of two cases. Hum Pathol
1985;16(8):851–852.
84. Erozgen F, Ersoy YE, Akaydin M, et al. Corticosteroid
treatment and timing of surgery in idiopathic granulomatous
mastitis confusing with breast carcinoma. Breast Cancer Res
Treat 2010;123(2):447–452.
85. Akbulut S, Arikanoglu Z, Senol A, et al. Is methotrexate an ac-
ceptable treatment in the management of idiopathic granuloma-
tous mastitis? Arch Gynecol Obstet 2011;284(5):1189–1195.
86. Ahmed YS, Abd El Maksoud W. Evaluation of therapeutic
mammoplasty techniques in the surgical management of
female patients with idiopathic granulomatous mastitis with
mild to moderate inflammatory symptoms in terms of recur-
rence and patients’ satisfaction. Breast Dis 2016;36(1):37–45.
87. Joseph KA, Luu X, Mor A. Granulomatous mastitis: a
New York public hospital experience. Ann Surg Oncol
2014;21(13):4159–4163.
88. Akbulut S, Yilmaz D, Bakir S. Methotrexate in the man-
agement of idiopathic granulomatous mastitis: review
of 108 published cases and report of four cases. Breast J
2011;17(6):661–668.
89. Sakurai K, Fujisaki S, Enomoto K, Amano S, Sugitani
M. Evaluation of follow-up strategies for corticosteroid
therapy of idiopathic granulomatous mastitis. Surg Today
2011;41(3):333–337.
90. Kiyak G, Dumlu EG, Kilinc I, et al. Management of idio-
pathic granulomatous mastitis: dilemmas in diagnosis and
treatment. BMC Surg 2014;14(1):66.
91. Asoglu O, Ozmen V, Karanlik H, et al. Feasibility of surgical
management in patients with granulomatous mastitis. Breast
J 2005;11(2):108–114.
92. Azlina AF, Ariza Z, Arni T, Hisham AN. Chronic granulo-
matous mastitis: diagnostic and therapeutic considerations.
World J Surg 2003;27(5):515–518.
93. Li S, Grant CS, Degnim A, Donohue J. Surgical manage-
ment of recurrent subareolar breast abscesses: Mayo Clinic
experience. Am J Surg 2006;192(4):528–529.
94. Patel RA, Strickland P, Sankara IR, Pinkston G, Many
W Jr, Rodriguez M. Idiopathic granulomatous mastitis:
case reports and review of literature. J Gen Intern Med
2010;25(3):270–273.
95. Elzahaby IA, Khater A, Fathi A, et al. Etiologic revelation and
outcome of the surgical management of idiopathic granulo-
matous mastitis; an Egyptian centre experience. Breast Dis
2016;36(4):115–122.
96. Yau FM, Macadam SA, Kuusk U, Nimmo M, Van Laeken
N. The surgical management of granulomatous mastitis. Ann
Plast Surg 2010;64(1):9–16.
97. Chai FY. Granulomatous mastitis: more data is needed.
Breast 2016;26:148.
98. Salehi M, Salehi H, Moafi M, et al. Comparison of the
effect of surgical and medical therapy for the treatment
of idiopathic granulomatous mastitis. J Res Med Sci
2014;19(suppl 1):S5–S8.
99. Yabanoğlu H, Çolakoğlu T, Belli S, et al. A comparative
study of conservative versus surgical treatment protocols for
77 patients with idiopathic granulomatous mastitis. Breast J
2015;21(4):363–369.
TM