ORIGINAL ARTICLE

Classification of Inflammatory Breast Disorders and Step

by Step Diagnosis
Rasha Mohamed Kamal, MD, Soha Talaat Hamed, MD, and
Dorria Saleh Salem, MD
Woman Imaging Unit, Radiodiagnosis Department, Kasr El Aini, Cairo University Hospitals,
Cairo, Egypt

n Abstract: In this study, the authors proposed a classification of inflammatory breast disorders based on which a
practical systematic scheme in diagnosis was applied aiming to differentiate simple forms of mastitis from more complicated
and malignant forms. The study population included 197 female patients who were clinically or pathologically diagnosed as
having mastitis. All patients underwent Ultrasound examination. Mammography was performed for 133 ⁄ 197 cases. Cases
of simple mastitis and periductal mastitis were followed up to ensure complete resolution. Abscess cavities and postopera-
tive collections were drained. Other cases were biopsied to confirm diagnosis and were managed accordingly by their treat-
ing physicians. Statistical analysis was performed by the Statistical Package for Social Science. Nominal Data were
expressed as frequency and relative frequencies (percentage). Ultrasound and Mammography categorical results were
compared using the Pearson Chi Square and Fisher’s exact test. Patients were classified into three groups; infectious, non-
infectious and malignant mastitis. Simple and malignant forms of mastitis showed many signs in common. The presence of
ill defined collections and abscess cavities on ultrasound favored simple over malignant forms of mastitis while extensive
skin thickening and infiltrated malignant nodes favored malignant forms. Interstitial edema, edematous fat lobules, abscess
cavities, skin thickening seen on ultrasound examination were significantly lower in noninfectious than simple and malignant
mastitis. Mammography signs were less discriminating. Diffuse skin thickening and increased density favored malignant
mastitis while dilated retro areolar ducts and characteristic calcification patterns favored noninfectious forms. Simple mastitis
showed nonspecific signs. Ultrasound examination in mastitis cases shows more specific signs in differentiating between
the three forms of mastitis and is useful in monitoring treatment, excluding complications and guide for interventional proce-
dures. Mammography should be performed whenever complicated, malignant and uncommon forms of mastitis are sus-
pected. n
Key Words: breast abscess, breast tuberculosis, duct ectasia, granulomatous mastitis, inflammatory carcinoma, mastitis,
plasma cell mastitis

R egardless of the type of breast problem, the goal

of the evaluation is to rule out cancer and address
the patient’s symptoms (1).
Mastitis is a frequently encountered complaint in
clinical practice. Although a significant problem espe-
cially amongst lactating women, yet, there remains a
paucity of scientific research into the anatomical, physi-
ological and pathological determinants of mastitis (2).
In general mastitis is inflammation of the breast
that may or may not be accompanied by infection.
The term mastitis is often used synonymous with
breast infection (3), but strictly speaking mastitis is
inflammation of the breast irrespective of the cause.
Address correspondence and reprint requests to: Rasha Mohamed
Kamal, MD, Woman Imaging Unit, Radiodiagnosis Department, Kasr El Aini,
Cairo University Hospitals, 52, Tayaran Street (Mahmoud Shaltoot), Nasr
City, Cairo, Egypt, or e-mail: rashaakamal@hotmail.com.
DOI: 10.1111/j.1524-4741.2009.00740.x
 2009 Wiley Periodicals, Inc., 1075-122X/09
The Breast Journal, Volume 15 Number 4, 2009 367–380
Having mastitis does not raise the woman’s risk of
developing breast cancer. However, an uncommon
type of breast cancer known as the inflammatory
breast cancer (IBC) has symptoms that are similar to
mastitis and can be sometimes mistaken as infection
(4).
Complete resolution is usually the rule in most
cases of mastitis If after ten days of antibiotics symp-
toms of mastitis do not dissipate, inflammatory carci-
noma should be ruled out and a biopsy should be
performed.
OBJECTIVE
In this study, we proposed a new classification of
inflammatory breast disorders based on which a prac-
tical systematic scheme in diagnosis was applied
aiming to differentiate simple forms of mastitis from
more complicated and malignant forms.
368 • kamal et al.
SUBJECTS AND METHODS
Subjects
The study population included 197 female patients
who were clinically or pathologically diagnosed as
having mastitis. Their ages ranged from 14 to 67 years
(average age: 39.8 years). They were referred from the
surgical outpatient’s clinics and wards. Patients were
classified into three main groups:
• GROUP 1: Infectious mastitis
• GROUP 2: Noninfectious mastitis
• GROUP 3: Malignant mastitis
Infectious mastitis (IM) encompasses breast specific
and nonspecific forms of infections whether primary
or complicating already present breast pathologies.
Noninfectious forms of mastitis (NIM) encompasses
another group of aseptic or chemical inflammatory
breast disorders.
The third group of mastitis, which is the most seri-
ous form of mastitis, is the malignant mastitis (MM)
usually accompanying the inflammatory breast carci-
noma or the very rare form of malignant breast
abscess.
Methods
All patients were subjected to complete medical his-
tory and full clinical examination by their referring
physicians.
Ultrasound examination was performed for all
cases (n = 197) using, a 7–13 MHz linear array trans-
ducer [Ultramark Philips HDI 5000 (Advanced Tech-
nology Laboratories (ATL), Best, The Netherlands) or
General Electric Logiq 7 machines (GE Healthcare,
Tokyo, Japan)]. The ultrasound report should confirm
or exclude the presence of:
• echogenic edematous fat lobules
• interstitial edema
• ill defined collections
• retro areolar duct system dilatation
• thickened skin (>2 mm)
• mass lesions (comment on their size, site, length ⁄
depth ratio, location, margin and echogenicity)
• cysts (comment on their size, number, location, wall
thickness and echogenic nature of their contents
• abscess cavities (diagnosed when a well defined
hypoechoic lesion with floating internal particles
was seen showing some degree of acoustic
enhancement identifying its cystic nature)
• fistulous tracts
• lymphnode enlargement (comment on their size
and hilum condition)
• presence or absence of internal vascularity
Mammography was performed for 133 ⁄ 197 cases
using Siemens Mammomat conventional mammo-
graphy machine (Siemens Mammomat, Erlangen,
Germany) or GE Senographe 2000 Full Field Digital
Mammography Machine (GE Senographe, Buc, Ver-
saie, France). Standard craniocaudal and mediolateral
oblique views were obtained; with the axilla included
in the latter. The mammogram report should confirm
or exclude the presence of:
• mass lesions
• focal asymmetric breast densities
• diffuse increased breast density
• retroareolar duct system dilatation
• thick skin (>2 mm)
• calcification (comment on the size, nature, num-
ber and distribution)
• lymphnodes
Magnetic resonance imaging (MRI) examination was
performed for selected cases (n = 6 ⁄ 197) using a 1.5
Tesla Philips Intera machine, Eindhoven Area, Nether-
lands. These cases included two cases of simple mastitis,
three IBCs and one case of tuberculous mastitis.
Cases of simple mastitis and active periductal mas-
titis were followed up on a repeated ten days basis to
ensure complete resolution. Abscess cavities (40 cases),
infected galactoceles (four cases) and postoperative
collections (15 cases) were drained and cytological
assessment was performed. Cytological assessment of
sinus tracts discharge and complicated cysts aspirates
were performed for 15 cases. Other cases were biop-
sied (fine needle biopsy for 17 cases, and core biopsy
for 18 cases) to confirm diagnosis and were managed
accordingly by their treating physicians.
Ultrasound and mammography findings were tabu-
lated. Statistical analysis was performed by the SPSS sta-
tistical program (Statistical Package for Social Science,
SPSS, Chicago, IL) using a Compaq Presario HP
Computer. Nominal Data were expressed as frequency
and relative frequencies (percentage). Ultrasound and
Mammography categorical results were compared using
the Pearson Chi Square and Fisher’s exact test. Probability
values (p-value) less than 0.05 were considered significant.
RESULTS
One hundred and ninety seven female patients clini-
cally or pathologically diagnosed as having mastitis

were included in the study. We classified them into
three main groups of mastitis according to a new pro-
posed classification by the authors (Fig. 1) based on
which a diagnostic pyramidal scheme was built up
starting up the diagnostic pyramid were diagnosis is
confirmed going down to the different imaging guided
interventional procedures in mastitis (Fig. 2). The fre-
quency and relative frequencies (percentage) of cases
included under each pathological type are listed in
Table 1.
Group I comprised the higher frequency of mastitis
cases (n = 132, 67%) (Fig. 3) showing predominance
of complicated forms of mastitis comprising 79
(40.1%) patients (Fig. 4). Fifty (37.9%) of Group I
patients were lactating. Group II included 54 (27.4%)
with predominance of periductal mastitis ⁄ duct ectasia
comprising 34 (17.3%) cases. Group III included 11
(5.6%) cases of malignant mastitis.
Ages of these patients ranged from 14 to 67 years
(average age: 39.8 years). In Group I, II 87 ⁄ 132
(65.9%) and 26 ⁄ 54 (48.1%) consecutively were below
40 years. All patients in Group III were above
40 years.
Table 2 shows the correlation of the ultrasound
findings in the three study groups stressing on the sig-
nificance of each sign in differentiating between the
different pathological entities of mastitis. The presence
of acute inflammatory signs on ultrasound (edema-
tous, echogenic fat lobules and interstitial edema)
Figure 1. New proposed classification of
mastitis.
Inflammatory Breast Disorders • 369
Figure 2. Diagnostic pyramid scheme.
could significantly differentiate infectious and malig-
nant forms of mastitis (p < 0.00) from noninfectious
forms but could not differentiate infectious from
malignant forms (Fig. 5). The presence of ill defined
collections (Fig. 6) and abscess cavities (Fig. 7) in
infectious mastitis (p = 0.001 and 0.026) and the
absence of malignant axillary nodes (p = 0.05) are sig-
nificant differentiating signs between infectious and
malignant mastitis. Mass lesions favor noninfectious
and malignant forms of mastitis (p = 0.038 and
p = 0.023) over infectious forms. Thickened skin was
significantly higher in malignant than infectious and
noninfectious mastitis (p = 0.01 and p < 0.05 respec-
tively) but no definite cutoff values could be calculated
due to the paucity of cases included in the study
(Fig. 8). Thick skin was also significantly higher in
infectious than noninfectious mastitis (p < 0.05).
Dilated ducts are significantly higher in noninfectious
370 • kamal et al.

Table 1. The Incidence of the Different Pathologi- tified between the three forms of mastitis apart from
cal Entities Included in the Study newly developed diffusely increased breast density
(p < 0.05) and marked skin thickening (p < 0.05)
Pathology Number Percentage
which were both significantly higher in malignant
Infectious 132 67 mastitis. These two signs were also higher in infec-
Simple mastitis 46 23.3
Lactational 29 14.7 tious than noninfectious forms (p = 0.001 and 0.012
Nonlactational 17 8.6 respectively). The presence of dilated retroareolar
Complicated mastitis 79 40.1
Abscess cavities 40 20.3
ducts and characteristic calcification patterns although
Infected cysts 18 9.2 seen in some cases of tuberculous mastitis (Fig. 12)
Infected galactoceles 4 2.
favor noninfectious forms of mastitis (Fig. 13).
Infected postinterventional collections 15 7.6
Infected hamartoma 1 0.5
Infected haematoma 1 0.5
Specific mastitis (TB) 7 3.6
DISCUSSION
Noninfectious mastitis 54 27.4
Periduct mastitis ⁄ duct ectasia 34 17.3 In general mastitis is inflammation of the breast
Plasma cell mastitis 6 3.1
Granulomatous mastitis 2 1 that may or may not be accompanied by infection.
Diabetic mastopathy 2 1 Mastitis can occur in all populations, whether or
Secondary mastitis 10 5
Scleroderma 2 1
not breastfeeding is the norm. The reported incidence
Sarcoidosis 1 0.5 varies from a few to 33% of lactating women, and
Post-traumatic fat necrosis 2 1
less than 10% in nonlactating ones (3).
Foreign body 5 2.5
Malignant mastitis 11 5.6 We classified mastitis into three main types; the
Inflammatory carcinoma 9 4.6 infectious, noninfectious and malignant mastitis.
Malignant breast abscess 2 1
Total 197 100 Infectious mastitis encompasses breast specific and
nonspecific forms of infections whether primary or
complicating already present breast pathologies. This
than infectious and malignant mastitis (p < 0.05) form of mastitis is more common during the child
(Fig. 9). Counts are only high in infectious mastitis in bearing period especially during lactation. Patients in
cases of lactational mastitis were ducts are physiologi- this group usually present with fulminant inflamma-
cally dilated. tory manifestations and are usually treated with anti-
Nine fistulous tracts with draining skin sinuses biotics, hot fomentations and various breast drainage
could be traced on ultrasound examination. Seven procedures. This group showed the highest incidence
were seen draining neglected abscess cavities. A non- in our study including 132 ⁄ 197 (67%) patients;
discharging fistulous tract was traced deep to the 84 ⁄ 132 (63.6%) of whom are below 40 years and
sternoclavicular joint in a tuberculous patient (Fig. 10) 50 ⁄ 132 (37.9%) of whom were lactating.
and another one was seen draining a malignant Noninfectious mastitis encompasses another group
abscess (Fig. 11) and Table 3 shows the correlation of aseptic or chemical inflammatory breast disorders
between the mammography findings in the three study that do not necessarily occur during lactation, are not
groups. No significant differentiating signs were iden- necessarily accompanied by microbial infections and

Figure 3. Acute lactational mastitis. A 22-

year-old lactating female presenting with
acute inflammatory signs of the right breast.
Ultrasound examination (a) showed edema-
tous fat lobules delineated by interstitial
edema lines with marked overlying skin thick-
ening. The patient was asked to come for a
follow-up ultrasound examination on a
2 weeks basis (under antibiotic coverage)
until complete resolution of the inflammatory
(a) (b) process was assured (b).
 Inflammatory Breast Disorders • 371

Figure 4. Secondary infected cysts. A 35-

year-old female, with fibrocystic disease of
both breasts. She developed acute inflamma-
tory signs of the left breast. An oval shaped
left retro-areolar mass lesion was seen in the
MLO mammogram (a). On ultrasound exami-
nation multiple secondary infected thick
walled cyst (b) were detected with precipitat-
ing echogenic debris (arrows). (a) (b)

Table 2. Correlation Between Ultrasound Findings in the Three Study Groups of Mastitis

Pathology, n ⁄ 197 Infectious, n ⁄ 132 Noninfectious, n ⁄ 54 Malignant, n ⁄ 11 P1 P2 P3

Echogenic fat, 116 ⁄ 197 (58.88) 99 (75) 7 (12.96) 10 (90.9) 0.2 <0.00* <0.00*
Interstitial edema, 116 ⁄ 197 (58.88) 99 (75) 7 (12.96) 10 (90.9) 0.2 <0.00* <0.00*
Ill defined collections, 71 ⁄ 197 (36) 71 (53.78) 0 (0) 0 (0) 0.001* — <0.05*
Thick skin, 97 ⁄ 197 (49.23) 83 (62.87) 3 (5.55) 11 (100) 0.01* <0.05* <0.05*
Dilated ducts, 80 ⁄ 197 (40.6) 40 (30.3) 40 (74) 0 (0) 0.03* <0.05* <0.05*
Sinus tracts, 9 ⁄ 197 (4.56) 8 (6) 0 (0) 1 (9.1) 0.52 0.16 0.06*
Mass, 11 ⁄ 197(5.58) 2 (1.5) 5 (9.25) 4 (36.36) <0.05* 0.038* 0.023*
Cyst ⁄ abscess, 68 ⁄ 197 (34.5) 66 (50) 0 (0) 2 (18.18) 0.026* 0.001* <0.05*
LN <0.05* <0.05* 0.3
Nonspecific, 168 ⁄ 197(85.2) 116 (92.8) 52 (96.29) 0 (0)
Pathological, 22 ⁄ 197 (11.16) 9 (7.2) 2 (3.7) 11 (100)

Values in parentheses are expressed as percentages.

P1: correlation between infectious and malignant.
P2: correlation between noninfectious and malignant mastitis.
P3: correlation between infectious and noninfectious mastitis.

Figure 5. Two different cases; (a) infective

mastitis and (b) malignant mastitis (IBC)
showing the same ultrasound signs: edema-
tous fat lobules, interstitial edema and
marked overlying skin thickening. (a) (b)

thus do not usually present with fulminant inflamma- This group encompassed 54 (27.4%) patients; 28
tory signs and do not usually resolve with antibiotics. (51.9%) of whom were below 40 years. With the excep-
Microbial infection may trigger some forms as periduc- tion of periductal and plasma cell mastitis these cases
tal mastitis or complicate others as diabetic mastopathy. were diagnosed on revision of their biopsy specimens.
372 • kamal et al.
(a) (b)
(a)
(b) (c)
The third group of mastitis, which is the most seri-
ous form of mastitis, is the malignant mastitis (MM)
usually accompanying the inflammatory breast carci-
noma or the very rare form of malignant breast
abscess. This group encompassed 11 (5.6%) patients.
Again microbial infection may be identified in these
cases complicating an already present malignant
pathology. All eleven cases (100%) were above
40 years, and none were lactating.
Unfortunately, there is a paucity of references dis-
cussing inflammatory breast disorders. For this reason
Figure 6. Two different cases of infective
mastitis showing ill defined collections
(arrows) bisecting the planes between the
edematous fat lobules.
Figure 7. Multiple abscess cavities. Nonlac-
tating 38-year-old female patient, presenting
with a hard retroareolar mass lesion associ-
ated with markedly enlarged axillary lymph-
nodes, suspected clinically to be malignant.
Mammography (a) showed a large partially
obscured retroareolar mass lesion (circle).
On ultrasound (b, c), two abscess cavities
were identified; one (c) showing tendency to
ulcerate through the skin.
we thought of proposing a new classification of
inflammatory breast disorders based on which we
built our diagnostic pyramidal scheme. Going down
this pyramid step by step we will provide a clear
structured diagnostic scheme whenever we are faced
by a case of mastitis.
Starting high up in our diagnostic pyramid we will
start by confirming the diagnosis of simple mastitis.
Mastitis occurs usually as a potential complication of
lactation (3). Most cases of mastitis resolve, but if
symptoms are aggravated, imaging becomes necessary.

(a)
Figure 8. Inflammatory carcinoma of the right
breast. A 47-year-old female. She developed
fulminant acute inflammatory signs of the right
breast. Mammography examination showed
right breast coarse trabeculae, marked skin
thickening (a), and pathological right axillary
lymph nodes (b). On ultrasound examination
(c), there was marked diffuse skin thickening
(double head arrow), edematous fat lobules
and interstitial edema. Infiltrated left supracla-
vicular lymphnodes were also identified on
ultrasound examination (d). (c)
A few previous studies have reported various
sonography and mammography features of mastitis
(5,6). Described mammography findings are also
vague and nonspecific (5). In our study we diagnosed
simple IM whenever branching anechoic interstitial
edema lines delineating the intervening plains between
the edematous and echogenic fat lobules in a nonduc-
tal distribution are identified on ultrasound examina-
tion. Associated skin thickening and reactive lymph
node enlargement are common. It is true that this pic-
ture although specific in differentiating IM from NM
it is not as specific as it is in differentiating between
IM and MM as confirmed from our results. We con-
sidered the presence of ill defined collections, not
described in previous studies, seen bisecting the planes
in-between the fat lobules a prognostic sign of IM.
These collections were not identified in any MM and
not even in NM (p = 0.001, p < 0.05). In our study
mammography findings were not specific in differenti-
ating IM from other forms. Diffuse breast density
(p = 0.03) and skin thickening (p = 0.02) favor IBC
Inflammatory Breast Disorders • 373
(b)
(d)
but theses signs could not exclude infectious mastitis
as. As mentioned previously, IM usually occur in
young lactating women an age and a time when the
breasts are inherently dense and painful. So in the
acute stages of mastitis we would advise go for an
Ultrasound and not for a mammography examination.
Ultizsch, et al. advised that patients with severe
mastitis should be periodically assessed until clinical
symptoms have resolved (7). We strongly support this
advice and thus all simple mastitis patients were asked
to come for a short term follow up in order to Moni-
tor treatment. On a short term follow up basis we will
again say go for an Ultrasound and not for a mam-
mography examination.
Moving another step downwards, we have to
exclude underlying complications. Acute bacterial
mastitis either resolves under antibiotic therapy or
evolves towards a pyogenic abscess especially if treat-
ment is delayed or inadequate (7,8). We defined
abscess cavities when we identified fairly well defined
oval or rounded cavities with low level internal
374 • kamal et al.
(a)
(b) (c)
echoes, good through transmission of sound that may
be surrounded by a focal area of mastitis; a similar
picture to that described in previous studies (7,9). We
have reported a higher incidence of pyogenic abscess
cavities complicating mastitis (n: 40, 20.3%) than that
reported in previous studies ranging from 4.8% to
11% (7,10). We believe that mammography should
not be used in the diagnosis of abscess cavities as find-
ings are nonspecific. Compression could not be ade-
quately applied because breasts are painful adding to
the already increased breast density due to the ongo-
ing inflammatory process.
Infected cysts (n = 18, 9.2%) and infected post-
operative seromas (n = 15, 7.6%) gave a similar picture
to abscess cavities being thick walled and containing
low level echoes and fluid debris (11). No recurrent or
residual mass lesions could be identified on ultrasound
examination so these cases were asked to come for a
short term ultrasound follow up under an antibiotic
coverage to ensure complete resolution. The presence
of intra mammary complex cysts and abscess cavities
in our study favored IM over NM (p < 0.05) and MM
(p = 0.001).
Secondary infected galactoceles were identified in
four cases giving a picture similar to an abscess cavity
but taking a ductal distribution in a lactating female.
Another two cases had already palpable identified
mass lesions that showed recent rapid increase in size
Figure 9. Periductal mastitis complicated with
breast abscess. A 45-year-old female com-
plaining of right paraareolar inflammatory
signs. (a) Ultrasound examination showed
mildly dilated retro areolar ducts with inspis-
sated secretions. Her symptoms were aggra-
vated with development of a left retro areolar
palpable mass lesion. A newly developed left
retro areolar mass lesion (b) was detected.
An underlying abscess cavity was identified
on ultrasound examination (c).
being associated with acute inflammatory signs. One
of these lesions followed breast traumatic insult.
Biopsy of these lesions revealed secondary infected
hamartoma and secondary infected post-traumatic
haematoma.
Nine (4.56%) fistulous tracts presenting by dis-
charging skin sinuses were identified in our study,
seven of which complicating inadequately treated,
nondrained abscess cavities, one complicating Tuber-
culosis infection and another complicating a malignant
breast abscess. Fistulous tracts reported by Almasad
JK, complicated similar pathologies in his study in
addition to periductal and granulomatous mastitis
(12). Some authors prefer to surgically remove adja-
cent mammary ducts in association with abscess cavi-
ties to prevent fistulas and recurrence (13).
We add our voice to the studies calling for an ultra-
sound examination to identify underlying abscess cavi-
ties being a unique means for evaluating the extent,
size, site and internal characteristics of abscess cavities
(8,14). We also think ultrasound is superior to mam-
mography in assessing secondary infected intra mam-
mary cysts, galactoceles, mastectomy and postbiopsy
beds. So, if abscess development or secondary infected
breast pathologies are suspected we would again
advise for an ultrasound and not for a mammography
examination unless suspicious associated mass lesions
are identified on ultrasound examination.

(a)
Figure 10. Healed Tuberculous mastitis with
persistent nondischarging skin sinus. A 60-
year-old female patient with a healed tuber-
culous calcified lesion in the right breast
apparent on both mammography (a) and
ultrasound (b) examinations. A contralateral
fistulous tract was traced by ultrasound
examination (c) from a nondischarging skin
sinus deep to the left sternoclavicular joint. (c)
Moving a further step downwards we have to
Exclude unusual forms of breast infections. Organisms
that are commonly responsible for mastitis include
staphylococcus and streptococcus species. Other rare
microorganisms can also infect the breast as salmo-
nella, fungi and mycobacterium tuberculosis. Breast
tuberculosis is a rare form of tuberculosis (15). The
significance of breast tuberculosis is due to rare occur-
rence and mistaken identity with breast cancer and
pyogenic breast abscess. Breast tuberculosis has no
defined clinical features and radiological imaging is
usually not diagnostic. Diagnosis is mainly based on
identification of typical histological features of the
tubercle bacilli under microscopy (16). In our study,
we encountered seven cases (3.6%) of breast tubercu-
losis; that were mainly diagnosed after biopsy. Tuber-
culosis is classified as primary when the breast lesion
is the only manifestation and secondary when there is
demonstrable focus elsewhere (17,18). Mammography
has been extensively explored for the diagnosis of
breast tuberculosis but of no avail (16). It is true that
the Ultrasound picture is as well not specific but at
least it can define the extent of the lesion rather than
diagnose, can characterize the breast lesions and is a
Inflammatory Breast Disorders • 375
(b)
more suitable examination as most patients with
breast tuberculosis are young 21–30 years (19). MRI
was performed in only one case of tuberculous masti-
tis with a considerably palpable breast lump eroding
the underlying ribs. MRI examination confirmed the
benign and cystic nature of the identified mass lesion
but its tuberculous nature was only confirmed after
biopsy.
So when breast tuberculosis is suspected we will
advise for ultrasound as well as mammography exami-
nations; as most tuberculous lesions are mistaken clin-
ically for malignancy. It is also sometimes essential to
identify characteristic calcification patterns. We will
also advise for other imaging modalities to identify a
primary tuberculous focus.
Moving another step downwards, we have to
exclude noninfectious forms of mastitis. The common-
est form of NM in our study was the periductal masti-
tis ⁄ duct ectasia constitituting 34 (17.3%) of mastitis
cases an incidence similar to that reported in other
studies (20,21).
Many terms, including duct ectasia, periductal mas-
titis, and plasma cell mastitis have been used in con-
nection with a variety of clinical conditions associated
376 • kamal et al.

(b)

Figure 11. Malignant breast abscess. A 65-

year-old female presenting with a palpable
mass lesion and discharging sinus in the
right axilla. Her mammogram (a) showed
(a) thickened para areolar skin and an apparent
pathological right axillary lymphnode. On
ultrasound examination (b, c) a deeply
seated right axillary mass lesion showing
areas of breaking down was detected. A fis-
tulous tract was traced by ultrasound exami-
nation, seen draining through the clinically
detected skin sinus. A malignant breast
(c) abscess was diagnosed on biopsy.

Table 3. Correlation Between Mammography Findings in the Three Study Groups of Mastitis

Pathology, n ⁄ 197 Infectious, n ⁄ 74 Noninfectious, n ⁄ 48 Malignant, n ⁄ 11 P1 P2 P3

Normal, 35 ⁄ 133 (26.3) 19 (25.67) 14 (29.16) 2 (18.18) 0.4 0.3 0.4

Dilated ducts, 23 ⁄ 133 (17.29) 3 (4) 20 (41.66) 0 (0) 0.6 0.006* <0.05*
Calcification, 13 ⁄ 133 (9.77) 3 (4) 10 (20.8) 0 (0) 0.6 0.09* 0.004*
Mass lesion, 37 ⁄ 133 (27.8) 25 (33.78) 9 (18.75) 3 (27.27) 0.4 0.3 0.05*
Focal density, 27 ⁄ 133 (20.3) 14 (18.9) 11 (22.9) 2 (18.18) 0.6 0.5 0.5
Diffuse density, 22 ⁄ 133 (16.54) 16 (21.6) 0 (0) 6 (54.54) 0.03* <0.05* 0.001*
Thick skin, 23 ⁄ 133 (17.29) 15 (20.27) 2 (4.16) 6 (54.54) 0.02* <0.05* 0.012*

Values in parentheses are expressed as percentages.

P1: correlation between infectious and malignant.
P2: correlation between noninfectious and malignant mastitis.
P3: correlation between infectious and noninfectious mastitis.
*Values less than 0.05 are statistically significant.

with nipple discharge, nonpuerperal sepsis and nipple
retraction a; a picture that mimics malignancy (22).
Duct ectasia ⁄ periductal mastitis is a benign disease
complex of uncertain etiology (23). The concept of
duct ectasia ⁄ periductal mastitis was challenged by
some authors who stated that they are different stages
in one disease process (24). We diagnosed duct ecta-
sia ⁄ periductal mastitis when we saw tubular anechoic
structures or ducts filled with debris and there may be
associated nipple discharge in agreement with previous
studies (25). On the other hand, plasma cell mastitis is
an aseptic inflammation of the breast. It is thought
that extravasation of intraductal secretions rich in
fatty acids causes chemical mastitis in the periductal
tissues with marked plasma cell infiltration (26).
Plasma cell mastitis was mainly diagnosed on mam-
mography. Large rod like branching calcifications,
usually >1 mm, with lucent centers are pathogno-
monic as described in previous studies (27). Calcifica-
tions are usually widespread and bilateral (28).
Idiopathic granulomatous mastitis is a rare disease
of the breast that clinically and radiologically mimics
breast carcinoma (29,30). Patients frequently have had
recent child birth. Biopsy usually reveal granuloma-
tous inflammation centered mainly on breast lobules.
The two cases encountered in the study presented by

(a)
Figure 12. Tuberculous Mastitis. A 50-year-
old female patient complaining of palpable
intra mammary, axillary and cervical mass
lesions. Large caseated intra mammary (a)
and axillary (b) lymphnodes were detected
on ultrasound examination. Enlarged calcified
upper (c) and lower (d) right deep cervical
lymphnodes are seen on postcontrast CT
examination of the neck (arrows). (c)
(a) (b)
Figure 13. A case of plasma cell mastitis showing typical ductal
calcifications (a) with central lucency on magnified view (b).
breast masses with inflammation of the overlying skin.
Dense calcification was detected in the mammography
examination of both cases and biopsy was essential to
diagnose both cases.
Diabetic mastopathy is also a source of confusion
with breast cancer. The association between mastopa-
thy and type 1 diabetes mellitus of long duration has
been reported (31). These patients present with rock
hard painless breast masses, a picture encountered in
only one of our cases with an associated breast carci-
noma. The other case presented with fulminant
inflammatory breast symptoms. Again biopsy was
Inflammatory Breast Disorders • 377
(b)
(d)
essential to make a diagnosis in both cases, showing
fibrosis and dense B-lymphocyte infiltration around
the lobules, a picture described by Yajimi et al. in
their study (32).
Sarcoid involvement of the breast is uncommon. It
is again difficult to differentiate breast sarcoidosis
from malignant breast lesions. therefore biopsy of all
suspicious lesions is essential (33). The single case of
sarcoidosis included in the study presented with exten-
sive cervical and left axillary lymphadenopathy and a
tender hard left upper outer quadrant mass lesion that
resolved under corticosteroid therapy. The mammog-
raphy picture was nonspecific with focal area of
increased left breast density. On ultrasound examina-
tion a corresponding ill defined area of focal mastitis
associated with markedly enlarged axillary and intra-
mammary lymphnodes having bizarre shaped eccentric
hila.
Dermatomyositis is a rare multisystem autoimmune
collagen vascular disorder of adults and children of
unknown etiology that primarily affects skin and ske-
letal muscle (34). The two cases of dermatomyositis
were included in the study; one of whom was of the
‘‘amyopathic variety’’ with no muscle involvement.
They presented with dermal calcifications proved after
skin biopsy to be calcified fibrolipomata.
Another important form of noninfectious mastitis is
the postconservative therapy irradiated breast. Diffuse
378 • kamal et al.
edema pattern with marked skin thickening and coars-
ened breast trabeculae are common mammography
findings in an irradiated breast. On ultrasound a pic-
ture similar to that described for simple mastitis is
usually encountered but again lacking the previously
described ill defined collections. These changes can
mask underlying residual or recurrent malignant
lesions. Post-irradiation changes can be confirmed by
MRI examination showing no underlying mass lesions
or by resolving symptoms on serial follow up exami-
nations.
Evaluating the different pathologies included under
the noninfective mastitis group, they did not have
much in common other than a clinical and imaging
presentation that mimicked malignancy and diagnosis
was only confirmed on biopsy. The presence of acute
inflammatory signs and skin thickening favored the
infective and malignant forms of mastitis. The pres-
ence of dilated ducts was significantly higher than the
other two groups (p < 0.05). The presence of mass
lesions was significantly higher than IM (p < 0.023)
and lower than the malignant variety (p = 0.038). So
when these cases are encountered we will say go for
ultrasonography and mammography examinations and
confirm by a biopsy to exclude malignant pathology.
Moving another step downwards we reach the most
serious form of mastitis; the malignant variety. Given
the grave prognosis for patients with inflammatory
breast cancer, diagnosis must not be delayed. If after
7–10 days of therapy inflammatory symptoms do not
dissipate, we should insist on a biopsy. Breast inflam-
mation in older nonlactating women should always
raise suspicion (35). All patients with malignant masti-
tis were nonlactating, above 40 years of age. Mam-
mography findings varied from ill defined dense
poorly demarcated areas of diffuse edema with exten-
sive skin thickening or mammography did not reveal
anything, since inflammatory symptoms lack specific-
ity as proved in previous studies (36). Findings in our
study go in agreement with the more detailed picture
described by Bilgen et al. as skin thickening, trabecu-
lar coarsening, and increased density with or without
mass lesions, microcalcifications and axillary lymph-
adenopathy (37). On US examination skin thickening
and parenchymal echogenicity changes owing to
lymph edema, were also detected in our study; a pic-
ture that did not help in the differentiation from acute
infective mastitis. Metastatic enlarged lymphnodes,
previously described by Bilgen et al. with eccentric or
absent hila and with a long to short axis ratio of less
than 1.5 were significantly higher in IBC (p < 0.05); a
sign which favored its diagnosis (37). Ill defined
collections detected in IM were never detected in
malignant mastitis. Comparing US to mammography,
US was superior to mammography in detection of skin
thickening. Mass lesions were also detected frequently
on US.
The presentation of malignancy as a breast abscess is
well described (38). Capellani et al. reported the fourth
worldwide squamous cell carcinoma presenting as an
abscess (39). The rate of associated malignancies with
breast abscess as reported in previous studies (40) is
very low. We reported two cases of malignant breast
abscesses in our study both diagnosed by biopsy. One
case gave a typical picture of a malignant breast abscess
with a markedly thickened wall, a picture which can be
confused with that of a chronic breast abscess. Associ-
ated typical pathological lymphnodes, with infiltrated
or eccentric hila, should favor a malignant pathology.
The other presented with an axillary discharging fistu-
lous tract seen draining an axillary abscess cavity with
an underlying spiculated outlined mass lesion. Both
cases were confirmed by excisional biopsy.
Any nonlactating female patient presenting with
inflammatory breast symptoms that fail to respond to
antibiotic therapy should be advised to go for both a
mammography and ultrasound examination immedi-
ately followed by a punch biopsy from the skin and
aspiration from the subdermal lymphatics to exclude
any possibility of IBC. The potential role of MRI in
the differentiation between benign and malignant
forms of mastitis has to be more deeply studied. We
performed MRI examination for only three cases of
inflammatory carcinoma and two cases with diffuse
mastitis inorder to confirm or exclude underlying mass
lesions. No significant change in the detected pattern
of enhancement could be used in the differentiating
between either forms and considerable overlap existed
between both of them. In a study performed by
Riebera et al. they concluded that while breast MRI
cannot currently be used definitively to distinguish
inflammatory carcinoma from mastitis, the differences
in dynamic enhancement may prove to be useful in
follow-up of presumed mastitis in problematic cases.
If after biopsy the diagnosis remains unclear, breast
MR may help to demonstrate the success of the antibi-
otic treatment and thus diagnose coexisting or con-
founding inflammatory carcinoma (41).
Reaching the last step in our diagnostic pyramids,
management should progress. Simple mastitis and

active periductal mastitis cases were asked to come for
a follow up ultrasound examination on a repeated ten
days basis until complete resolution was ensured.
Abscess cavities, infected galactoceles and postopera-
tive collections were referred back to their treating
physicians and were surgically drained. Cytological
assessment of the drained lesions was performed to
exclude the presence of any malignant cells. They were
also sent to perform a postoperative ultrasound exami-
nation to exclude recollections. Cytological assessment
of sinus tracts discharge and ultrasound guided aspi-
rates of complicated cysts were performed for 15 cases.
Other cases were biopsied (fine needle for 17 lesions
and core biopsy for another 18 lesions) under
ultrasound guidance to confirm diagnosis and were
managed accordingly by their treating physicians.
CONCLUSION
Ultrasound examination in mastitis cases shows
more specific signs in differentiating between the three
forms of mastitis and is useful in monitoring treat-
ment, excluding complications and guide for interven-
tional procedures. Mammography should be
performed whenever complicated, malignant and
uncommon forms of mastitis are suspected.
REFERENCES
1. Morrow M. Evaluation of common breast problems. Am Fam
Physician 2000;61:2371–85.
2. Fetherston C. Mastitis in lactating women: physiology and
pathology. Breastfeed 2001;9:21.
3. Inch S, Von Xylander S. Mastitis: Causes and Management.
Geneva: WHO, Department of Child and Adolescent Health and
Development, 2000.
4. Guray MS. Benign breast diseases: classification, diagnosis
and management. Oncologist 2006;11:435–49.
5. Gupta R, Behbehani A, Chisti F, Alsaleh M. The role of ultra-
sound in the management of inflammatory disorders of the breast.
Medical Principles and Practice 1999;8:51–7.
6. Thirumalaikumar S, Kommu S. Aspiration of breast abscesses.
Emerg Med J 2004;21:333–4.
7. Ulitzsch D, Nyman M, Carlson RA. Breast abscess in lactat-
ing women: US guided treatment. Radiology 2004;232:904–9.
8. Leborne F, Lebourne F. Treatment of breast abscesses with
sonographically guided aspiration, irrigation, and instillation of
antiobiotics. AJR 2003;181:1089–91.
9. Hook GW, Ikeda DM. Drainage without indwelling catheter
placement. Radiology 1999;213:579–82.
10. Bland KI, Copeland EM. The Breast: Comprehensive Man-
agement of Benign and Malignant Disease, 2nd edn. Philadelphia,
PA: Saunders, 1998.
11. Berg WA, Campassi CL, Loffe OB. Cystic lesions of the
breast sonographic pathologic correlation. Radiology 2003;227:
183–91.
Inflammatory Breast Disorders • 379
12. Almasad JK. Mammary duct fistulae, classification and
management. ANZ Journal of Surgery 2006;76:149–52.
13. Mguid MM, Oler A, Numann PJ. Subareolar breast abscess:
the penultimate stage of the mammary duct associated inflammatory
disease sequence. In: Bland KI, Copeland EM, eds. The Breast:
Comprehensive Management of Benign and Malignant Disease, 2nd
edn. Philadelphia, PA: Saunders, 1998:93–102.
14. Tiu CM, Chiou HJ, Chou YH, et al. Sonographic features of
breast abscess with emphasis on the hypoechoic rim sign. Zhonghua
Yi Xue Za Zhi (Taipei) 2001;64:153–60.
15. Puneet MS, Satyendra KT, Ragini R, Singh S, Gupta SK,
Shukla VK. Breast tuberculosis: still common in India. Internet J
Trop Med 2005; 2 Number 2, ISSN 1540-2681.
16. Tewari M, Shukla HS. Breast tubercolosis, clinical features
and management. N Indian J Med Res 2005;122(2):103–10.
17. Mackeown KC, Wilkinson KW. Tuberculous disease of the
breast. Br J Surg 1952;39:420.
18. Vassilakos P. Tuberculosis of the breast: cytological findings
and fine needle aspiration. A case clinically and radiologically mim-
icking carcinoma. Acta Cytol 1973;17:160–5.
19. Popli MB. Pictorial essay; tuberculosis of the breast. Indian
J Radiol Imag 1999;9:127–32.
20. Dixon JM, Dixon LM. Periductal mastitis ⁄ duct ectasia.
World J Surg 1989;13:715–20.
21. Dixon JM, Anderson TJ, Lumsden AB, Elton RA,
Robert MM, Forrest APM. Mammary duct ectasia. Br J Surg
1983;70:601–3.
22. Hughes LE. Non lactational inflammation and duct ectasia.
Br Med Bulletin 1991;47:272–83.
23. Al Masad JK. Mammary duct ectasia and periductal masti-
tis. Saudi Med J 2001;22:1030–3.
24. Dixon JM, Ravisekar O, Chatty O, Anderson TJ. Periductal
mastitis and duct ectasia: different conditions with different etiol-
ogy. Br J Surg 1996;83:820–2.
25. Weinstein SP, Conant E, Orel SG, Zuckerman JA, Bellah R.
Spectrum of ultrasound findings in pediatrics and adolescent patients
with palpable breast masses. Radiographics 2000;20:1613–21.
26. Manzulla E, Sneigo N, Singletary SE. Diagnosis of breast
mass in young woman. Primary care and cancer 1998;18, ISSN:
0743-8176.
27. Heywang SH, Dershaw DS. Inflammatory breast conditions.
In: Heywang-Kobrunner Sylvia, ed. Diagnostic Breast Imaging:
Mammography, Sonography, MRI and Interventional Procedures.
Thieme 2nd edn. Amazon: Thieme Medical Pub, 2001:236, Chapter
13.
28. Stephans TW, Whitman GJ. Breast benign calcifications.
eMedicine 2005. eMedicine specialities>Radiology>Breast.
29. Ahmed R, Sultan F. Granulomatous mastitis, a review of 14
cases. Y Ayub Med Coll Abboytabad 2006;18:52–4.
30. Bani H, Yaghan RJ, Matalka II, Shatnawi NJ. Idiopathic
granulomatous mastitis, time to avoid unnecessary biopsies. Breast J
2004;10:318–22.
31. Mak CW, Chou CK, Chen SY, Lee PS, Chang JM. Diabetic
mastopathy, case report. BJR 2003;76:192–4.
32. Yajimi S, Fukotomi T, Akashi-Tanaka S, Nanasawa T,
Miyakawa K, Hasegawa T. Diabetic mastopathy, a case report
with reference to the findings of enhanced computed tomography.
Breast cancer 2001;8:246–9.
33. Lower EE, Hawkins HH, Baughman RP. Breast disease in
sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis 2001;18:301–6.
34. Feder JM, de Parades ES, Hogge JP, Wilken JJ. Unusual
breast lesions; radiologic pathologic correlation. Radiographics
1999;19:11–26.
380 • kamal et al.
35. Dahlbeck SW, Donnelly JF, Theriault RL. Differentiating
inflammatory breast cancer from acute mastitis. Am Fam Physician
1995;52:929–34.
36. Bland KI, Copeland EM. The Breast: Comprehensive Man-
agement of Benign and Malignant Diseases, 3rd edn. Philadelphia:
Saunders, 1991.
37. Bilgen IG, Ustin EE, Memis A. Inflammatory breast carci-
noma: mamographic, ultrasonographic, clinical and pathologic find-
ings in 142 cases. Radiology 2002;223:829–38.
38. Wong C, Wright C, Colclough A, Marsh S. Case report:
metaplastic carcinoma presenting as a breast abscess. Int Semin Surg
Oncol 2006;3:32.
39. Cappellani A, Di Vita M, Zanghi A, et al. A pure squamous
cell breast carcinoma presenting as a breast abscess: case report and
review of literature. Ann Ital Chir 2004;75:259–62.
40. Scott BG, Silberfein EJ, Pham HQ, et al. Rate of malignan-
cies in breast abscesses and argument for ultrasound drainage. Am J
Surg 2006;192:869–72.
41. Riebera A, Tomczak RJ, Mergo PJ, Wenzel V, Zeitler H,
Brambs HJ. MRI of the breast in the differential diagnosis of masti-
tis versus inflammatory carcinoma and follow-up. J Comput Assist
Tomogr 1997;21:128–32.