DOI:10.2214/AJR.07.2479
AJR 2008; 191:359-363
© American Roentgen Ray Society
Analysis of the Mammographic and Sonographic Features of Pseudoangiomatous Stromal Hyperplasia
Gormlaith C. Hargaden1,2,
Eren D. Yeh1,3,
Dianne Georgian-Smith1,3,
Richard H. Moore1,
Elizabeth A. Rafferty1,
Elkan F. Halpern4 and
Grace T. McKee5
1 Division of Breast Imaging, Department of Radiology, Avon Comprehensive Breast
Center, Massachusetts General Hospital, Boston, MA.
2 Present address: Breast Check, Mater Misericordiae University Hospital, 36
Eccles St., Dublin 7, Ireland.
3 Present address: Division of Breast Imaging, Department of Radiology, Harvard
Medical School, Brigham and Women's Hospital, Boston, MA.
4 Institute for Technology Assessment, Massachusetts General Hospital, Boston,
MA.
5 Department of Pathology, Massachusetts General Hospital, Boston, MA.
Received April 30, 2007;
accepted after revision February 5, 2008.
Address correspondence to G. C. Hargaden
(gormlaith.hargaden{at}breastcheck.ie).
Abstract
OBJECTIVE. The purpose of this study was to describe the imaging
findings in 149 patients with pseudoangiomatous stromal hyperplasia (PASH) who
had undergone at least 4 years of clinical follow-up for detection of
subsequent malignancy.
CONCLUSION. PASH is a common entity that presents with benign
imaging features without evidence of subsequent malignant potential. At our
institution, in the absence of suspicious features a diagnosis of PASH at core
biopsy is considered sufficient, and surgical excision has been obviated.
Keywords: breast • mammography • pseudoangiomatous stromal hyperplasia • sonography
Introduction
Pseudoangiomatous stromal hyperplasia (PASH) of the breast has been
diagnosed with increasing frequency. Vuitch et al.
[1] in 1986 conducted a study
with nine premenopausal patients with breast masses consisting histologically
of mammary stromal proliferations simulating vascular lesions. It is now
recognized that PASH may be found incidentally in as many as 25% of breast
biopsy specimens [2],
predominantly with a probable hormonal cause in premenopausal women. The
cytologic features of PASH are almost identical to those of fibroadenoma, the
predominant finding being sheets of benign ductal cells
[1]. The histologic appearance
is anastomosing slitlike empty spaces lined by spindle cells. This
characteristic is in contrast to vascular channels, which are lined by
endothelial cells and have RBCs within them. Because they are myofibroblasts,
the lining cells usually have a positive CD34 stain result. Dense collagenous
stroma separates the spaces. It is currently believed
[1,
3,
4] that PASH is hormonally
induced and often accompanied by benign epithelial proliferation in ducts and
lobules.
Although the mammographic and sonographic appearances of PASH have been
described, the series have been small, only seven patients in two studies
[5,
6] and 13 patients in a study
of sonographic findings conducted by Mercado et al.
[7]. Our purpose was to
describe the mammographic and sonographic appearances of PASH in 149 patients
treated at our institution over a 5-year period who underwent at least 4 years
of clinical follow-up for the detection of malignancy and recurrence.
Materials and Methods
The radiology and surgical pathology databases at our institution were
retrospectively reviewed, and cases with a diagnosis including PASH were
identified. Institutional review board approval was obtained. From
1996–2001, approximately 10,000 surgical breast biopsies were performed
at our institution. The first case of PASH identified in the database was
reported in 1996. Over a 5-year period (1996–2001), 178 patients (177
women, one man) with a pathologic diagnosis of PASH who had undergone imaging
within 1 year before the diagnosis was made were identified. The images of 29
patients could not be obtained. The mammographic and sonographic images of the
other 149 patients were reviewed by an experienced breast radiologist.
The mammographic findings were described according to BI-RADS as
circumscribed mass, focal asymmetric density, architectural distortion,
calcifications, or negative. The sonographic findings were described in terms
of mass shape, margins, echo pattern, and posterior acoustic features. In
addition, the history with regard to mode of presentation (screening or
palpable abnormality), menstrual status, and use of exogenous hormones was
noted. Whether the patient underwent core or excisional biopsy also was
documented. The patients underwent follow-up for 4 years to determine whether
carcinoma had developed at the site or whether PASH had recurred.
View larger version (15K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1 —Graph summarizes mammographic findings: normal, mass lesion
(circumscribed), focal asymmetric density, and incidental finding of
architectural distortion or calcifications. Light gray indicates screening
group; dark gray, clinical finding group.
|
|
Results
The patients were subdivided into two groups on the basis of initial mode
of presentation. The first group consisted of 59 patients in whom
abnormalities were found at routine screening. Thirty-eight patients in this
group also underwent sonography, and 16 (27% of the 59) were found to have
abnormalities. The second group consisted of 90 patients with 91 lesions
detected as palpable abnormalities at clinical examination. Three of these
patients had concurrent carcinoma in the opposite breast, and one patient had
a history of carcinoma in the same breast. Two patients were pregnant and did
not undergo mammography. Sonograms of a total of 71 patients were available
for review, and abnormalities were identified in 36 (51%) of the patients. In
total, 35 (23%) of the 149 patients had no abnormalities identified on images.
Core biopsies were performed on 20% of the patients (17% of the screening
group, 21% of the clinical finding group), and surgical excision was performed
on 89% (86% of the screening group, 91% of the clinical finding group).
The most common mammographic abnormalities
(Fig. 1) were a circumscribed
mass or asymmetric density. In the screening group we identified two cases of
architectural distortion and seven cases of calcifications. On mammography,
one of the two cases of architectural distortion exhibited increasing
architectural distortion at a previous surgical biopsy site, and pathologic
examination revealed previous surgical scar with incidental PASH. The
pathologic findings in the second case showed radial scar with incidental
PASH. Pathologic examination showed that three of the seven patients with
calcifications had ductal carcinoma in situ (DCIS), two had atypical ductal
hyperplasia, and two had fibrocystic changes. In all cases, associated
calcifications were found, and PASH was an incidental finding. In the clinical
finding group, 69% of the patients had no identifiable mammographic
abnormality; there were no cases of architectural distortion. The one patient
in whom calcifications were found had the biopsy findings of invasive ductal
carcinoma with DCIS and incidental PASH. In patients with sonographic
abnormalities (Table 1), the
most common finding was a well-circumscribed oval hypoechoic mass.
DCIS was found in three patients in the screening group and one patient in
the clinical finding group. All three cases of DCIS identified
mammographically had calcifications associated with DCIS, and PASH was an
incidental finding. There was one case of invasive carcinoma in each group. In
all of the cases of associated malignancy, PASH was an incidental finding.
Lobular carcinoma in situ was identified in one patient in the clinical
finding group. Interestingly, one patient in whom PASH was identified had a
low-grade phyllodes tumor in the contralateral breast. Of the 149 patients
with PASH, there were no (0%; 95% CI, 0–2.4%) subsequent cancers
detected at the site of PASH in the 4-year follow-up period. In the other
cases, after radiologic–pathologic correlation, PASH was determined by
the significant pathologic finding (Figs.
2 and
3) either alone or associated
with fibrocystic change, fibroadenoma, atypical ductal hyperplasia, or other
abnormalities, including sclerosis, atypical lobular hyperplasia, hamartoma,
radial scar, papilloma, and fibrosis. Most of the women were either
premenopausal (screening group, 51%; clinical finding group, 75%) or taking
hormonal replacement therapy (screening group, 25%; clinical finding group,
10%).
View larger version (25K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2 —Graph shows pathologic findings in screening group.
Pseudoangiomatous stromal hyperplasia was identified in all cases, either
alone or with invasive carcinoma, fibrocystic change, fibroadenoma, atypical
ductal hyperplasia, ductal carcinoma in situ, or another abnormality.
|
|
View larger version (26K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3 —Graph shows pathologic findings in clinical group.
Pseudoangiomatous stromal hyperplasia was identified in all cases either alone
or with invasive carcinoma, fibrocystic change, fibroadenoma, atypical ductal
hyperplasia, ductal carcinoma in situ, or another abnormality.
|
|
Discussion
Since PASH was first described in 1989, the pathologic features have been
well documented. There remains, however, little in the radiology literature
describing the imaging findings
[5–7].
Because PASH is identified in as many as 25% of biopsy specimens
[2], it is important to be
familiar with the radiologic findings so that unnecessary surgery can be
avoided. Initially at our institution patients with a diagnosis of PASH at
core biopsy underwent surgical excision; however, we now recognize that course
of management is unnecessary. If the imaging findings are concordant with the
pathologic findings and there are no suspicious features, return to routine
screening is recommended.
View larger version (72K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A —62-year-old postmenopausal woman not taking hormone
replacement therapy who underwent screening mammography. Screening mammogram
in extended craniocaudal (A) and mediolateral oblique (B) views
shows new focal asymmetric density (arrow) in left upper outer breast
compared with mammogram (not shown) obtained 3 years previously. It measures
approximately 2 cm, is low in x-ray attenuation, and has ill-defined margins.
Because it was new and has ill-defined margins, lesion is concerning for
malignancy, and biopsy was suggested. Results of needle localization and
surgical biopsy showed pseudoangiomatous stromal hyperplasia.
|
|
View larger version (81K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —62-year-old postmenopausal woman not taking hormone
replacement therapy who underwent screening mammography. Screening mammogram
in extended craniocaudal (A) and mediolateral oblique (B) views
shows new focal asymmetric density (arrow) in left upper outer breast
compared with mammogram (not shown) obtained 3 years previously. It measures
approximately 2 cm, is low in x-ray attenuation, and has ill-defined margins.
Because it was new and has ill-defined margins, lesion is concerning for
malignancy, and biopsy was suggested. Results of needle localization and
surgical biopsy showed pseudoangiomatous stromal hyperplasia.
|
|
View larger version (126K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4C —62-year-old postmenopausal woman not taking hormone
replacement therapy who underwent screening mammography. Photomicrograph shows
typical pseudoangiomatous stromal hyperplasia: anastomosing slitlike empty
spaces lined by flattened myofibroblasts without RBCs. (H and E)
|
|
In none of our 149 patients was carcinoma found at the same site as PASH in
at least 4 years of follow-up. Two patients had subsequent diagnoses of
malignancy in the opposite breast. The rate of recurrence of PASH after
excision has been described in the literature as ranging from 15% to 22%
[1,
4]; however, our recurrence
rate was not that high. Three (2%) of the patients had recurrence of PASH, and
the rest had no mammographic or clinical evidence of recurrence.
View larger version (71K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5A —53-year-old postmenopausal woman taking hormone replacement
therapy who has palpable lump. Craniocaudal (A) and mediolateral
oblique (B) mammograms of left breast show mass (arrow) in
retroareolar region, corresponding to patient's focal area of clinical
concern. Mass measures 3 cm and has partially well-circumscribed margins and
margins partially obscured by adjacent dense breast tissue.
|
|
View larger version (85K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B —53-year-old postmenopausal woman taking hormone replacement
therapy who has palpable lump. Craniocaudal (A) and mediolateral
oblique (B) mammograms of left breast show mass (arrow) in
retroareolar region, corresponding to patient's focal area of clinical
concern. Mass measures 3 cm and has partially well-circumscribed margins and
margins partially obscured by adjacent dense breast tissue.
|
|
View larger version (98K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5C —53-year-old postmenopausal woman taking hormone replacement
therapy who has palpable lump. Focal sonogram corresponding to A and
B shows 2.9-cm solid hypoechoic oval mass with smooth
well-circumscribed margins and unusually internal cystic spaces.
Sonographically guided core biopsy and pathologic examination revealed dense
sclerosis and features suggestive of pseudoangiomatous stromal hyperplasia.
Surgical biopsy revealed pseudoangiomatous stromal hyperplasia with associated
fibrocystic changes. Crosses delineate lesion.
|
|
View larger version (97K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5D —53-year-old postmenopausal woman taking hormone replacement
therapy who has palpable lump. High-power photomicrograph shows lesion more
cellular than that in Figures
4A,
4B, and
4C. Slitlike spaces are
collapsed or compressed, and many plump myofibroblasts are present. Dense
collagenous stroma separates spaces. (H and E)
|
|
In keeping with earlier descriptions
[5], the most common
mammographic abnormalities documented in our study were a circumscribed mass
and, less commonly, a focal asymmetric density (Figs.
4A,
4B,
4C,
5A,
5B,
5C,
5D). We found no spiculated
lesions, and architectural distortion and calcifications were uncommon and
unrelated to PASH. A large proportion (69%) of the patients with the clinical
presentation of PASH had no mammographic abnormalities. As one would expect
with a benign process, the sonographic appearance was most commonly a
well-circumscribed hypoechoic or isoechoic oval mass with enhanced through
transmission. Although they are nonspecific, these imaging findings have the
features of a benign process
[8], and thus the pathologic
diagnosis would be concordant with the imaging findings, making further
sampling unnecessary.
It is important to remember that a case with any suspicious features
warrants further sampling because PASH can coexist with a malignant process
and should not be accepted as a final diagnosis on the basis of core biopsy
findings alone. In a review of 200 consecutively collected breast specimens,
invasive adenocarcinoma was found in 10% of cases of PASH
[5,
6]. In our series, 4% of
patients had coexistent carcinoma at the site of PASH (all of these patients
had calcifications on mammography). Damiani et al.
[9] in 2002 studied five
malignant tumors of the breast that had an unusual pattern of spread within
anastomosing spaces similar to those described in PASH. Those authors
suggested that the spaces may represent part of a prelymphatic network and
thus be an unrecognized pathway of neoplastic spread. Although much work needs
to be done to substantiate this proposal, it may mean the presence of PASH
adjacent to a malignant process would have important clinical
implications.
The increasing proportion of biopsy cases in which PASH was diagnosed over
the period of this study is attributable to the familiarity of pathologists at
our institution with what was once thought to be an artifact within a
specimen. Although a large proportion of patients in both groups had PASH
alone (53 patients, 36%), one half of the cases were associated with
fibrocystic change (74 patients, 50%), and this lesion was the most common
associated finding in both groups of patients.
From its earliest descriptions
[1] in which mass lesions were
identified in nine premenopausal women, PASH has been thought to have a
hormonal basis. The stromal changes identified within PASH are similar to
breast changes during the luteal phase of menstruation
[3]. One of the patients
described by Powell et al. [4]
achieved temporary control of extensive PASH with hormonal manipulation. In
our series, most of the women were either premenopausal (screening group, 51%;
clinical finding group, 75%) or taking hormonal replacement therapy (screening
group, 25%; clinical finding group, 10%), which would be in keeping with the
finding by Powell et al. Interestingly, there were more postmenopausal women
not taking hormonal replacement therapy in the screening group (19% vs 8%),
which may mean there was insufficient hormonal influence to make the PASH
changes large enough to palpate. The only man in our study had gynecomastia in
association with PASH, again suggesting a hormonal basis because gynecomastia
is a response in the male breast to hormonal manipulation.
Angiosarcoma can be confused histologically with PASH. It is important to
differentiate the two lesions because of differences in prognosis and
treatment. PASH is benign, necessitating no further intervention, whereas
angiosarcoma is a tumor usually managed with wide local excision and
chemotherapy [10].
Angiosarcoma is highly vascular and bleeds easily, has branching blood vessels
lined by endothelial cells, and has no collagenous stroma.
PASH is a benign lesion believed to be hormonally induced and is not
thought to be associated with an increased incidence of malignancy. In our
series of 149 patients, PASH was an incidental finding in all cases associated
with malignancy, and no patient subsequently had carcinoma at the site of
PASH. Our data contained no evidence that PASH is related to subsequent
development of cancer; that is, PASH is not a high-risk lesion but a diagnosis
in itself. In the absence of suspicious imaging findings, it is highly
unlikely that associated undiagnosed malignancy is present.
A potential limitation of our study was the small number of patients with
malignancy. At our institution the finding of PASH in a pathologic specimen is
not always documented if there is a more important finding such as malignancy.
This factor may have been another limitation of our study because we might
have evaluated only a proportion of the cases in which PASH might have been
found.
As PASH is increasingly recognized as a pathologic entity, both alone and
in association with other findings, it is important that radiologists become
familiar with the various mammographic and sonographic appearances to better
advise management. Although there are no diagnostic imaging criteria, in our
study we found that when visible mammographically, PASH manifests most
commonly as a circumscribed mass or focal asymmetric density. On sonography it
has benign features, commonly an oval hypoechoic mass with enhanced through
transmission. At our institution, in the absence of suspicious features, if
sampling is adequate, a diagnosis of PASH at core biopsy is acceptable, and
surgical excision is obviated.
References
- Vuitch MF, Rosen PP, Erlandson RA. Pseudoangiomatous hyperplasia of
mammary stroma. Cancer Hum Pathol 1986;17
: 185-191[CrossRef]
- Ibrahim RE, Sciotto CG, Weidner N. Pseudoangiomatous hyperplasia of
mammary stroma. Cancer 1989;63
: 1154-1160[CrossRef][Medline]
- Longacre TA, Bartow SA. A correlative morphologic study of human
breast and endometrium in the menstrual cycle. Am J Surg
Pathol 1986; 10:382
-393[CrossRef][Medline]
- Powell MC, Cransor ML, Rosen PP. Pseudoangiomatous stromal
hyperplasia (PASH). Am J Surg Pathol1995; 19:270
-277[Medline]
- Polger MR, Denison CM, Lester S, Meyer JE. Pseudoangiomatous
stromal hyperplasia: mammographic and sonographic appearances.
AJR 1996; 166:349
-352[Abstract/Free Full Text]
- Cohen MA, Morris EA, Rosen PP, Dershaw DD, Liberman L, Abramson AF.
Pseudoangiomatous stromal hyperplasia: mammographic, sonographic, and clinical
patterns. Radiology 1996;198
: 117-120[Abstract/Free Full Text]
- Mercado CL, Naidrich SA, Hamele-Bena D, Fineberg SA, Buchbinder SS.
Pseudoangiomatous stromal hyperplasia of the breast: sonographic features with
histopathologic correlation. Breast J2004; 10:427
-432[CrossRef][Medline]
- Stavros AT. Breast ultrasound.
Philadelphia, PA: Lippincott Williams and Wilkins, 2004:445
-527
- Damiani S, Peterse JL, Eusebi V. Malignant neoplasms infiltrating
"pseudoangiomatous" stromal hyperplasia of the breast: an
unrecognized pathway of tumour spread. Histopathology2002; 41:208
-215[CrossRef][Medline]
- Johnson CM, Garquilo GA. Angiosarcoma of the breast: a case report
and literature review. Curr Surg 2002;59
: 490-494[CrossRef][Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
