| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Breast Imaging |
1 From the Departments of Radiology (C.W.P., S.A.F.) and Pathology, Anatomy and Cell Biology (J.P.P.), Jefferson Medical College and Thomas Jefferson University Hospital, 132 S 10th St, 7th Fl, Main Bldg, Philadelphia, PA 19107. From the 1997 RSNA scientific assembly. Received April 16, 1998; revision requested June 24; revision received August 5; accepted October 6. Address reprint requests to C.W.P.
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Serial mammograms obtained in 21 women with negative physical examination results and mammographically detected developing asymmetric breast tissue were reviewed, and findings were correlated with results of biopsy (n = 16), ultrasonography (US) (n = 8), and contrast material–enhanced magnetic resonance (MR) imaging (n = 3). Five patients who did not undergo biopsy were followed up for 13–84 months. Thirteen of 16 biopsy specimens were reviewed.
RESULTS: At the time of mammographic change, 12 patients without baseline asymmetric tissue had a mean age of 41.7 years and a mean size of asymmetric tissue of 2.4 cm. The mean age of nine patients with baseline asymmetric tissue was 46.9 years. In eight patients, the mean size increase was 2.5 cm. One patient showed increased tissue density but stable size. All US and MR images were negative. Pseudoangiomatous stromal hyperplasia was present in all 13 biopsy specimens reviewed and extensive in 12. No malignancies have been reported in five of the followed-up patients, and two have had continued enlargement of asymmetric tissue.
CONCLUSION: Pseudoangiomatous stromal hyperplasia is a common histopathologic finding in developing asymmetric breast tissue. Follow-up, rather than biopsy, is a management option if benign imaging and clinical criteria are met.
Index terms: Breast, biopsy, 00.1261, 00.1267 • Breast, diseases, 00.319, 00.721, 00.723 • Breast, MR, 00.121412, 00.121415, 00.12143 • Breast neoplasms, 00.319 • Breast, US, 00.1298
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Such distinctions provide better criteria for the recommendation of supplementary mammographic views and/or US (BI-RADS category 0, incomplete assessment), routine screening (BI-RADS category 1, normal), routine screening (BI-RADS category 2, benign), short interval follow-up (BI-RADS category 3, probably benign), and biopsy (BI-RADS categories 4, moderately suspicious abnormality, and 5, highly suspicious of malignancy).
Asymmetric breast tissue has been reported to be depicted on 3% of mammograms and is nearly always benign (3). Biopsy of a stable finding that fulfills the American College of Radiology's BI-RADS definition of asymmetric tissue is rarely indicated because malignancy has been found in only 3% of cases (3,4). Because some focal asymmetric densities may represent masses with borders that are either ill-defined or obscured by surrounding fibroglandular tissue rather than asymmetric tissue, further imaging evaluation of mammographic findings categorized as a focal asymmetric density may be necessary. In addition, a focal asymmetric density that corresponds to a palpable finding requires further evaluation (3).
Although soft-tissue asymmetry represents an uncommon manifestation in breast cancer, an increasing soft-tissue density may be of more concern. Among 300 nonpalpable cancers reported by Sickles (5), 243 (81%) manifested as masses or calcifications; 26 (9%), as architectural distortion; 19 (6%), as developing density; and eight (3%), as a soft-tissue "asymmetry." To show that benign asymmetric breast tissue may develop and change over time, we present a series of patients who had asymmetric breast tissue that either newly appeared, enlarged, or became more radiographically dense on serial mammograms.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The cases reviewed were those of the 21 patients who underwent biopsy or had 13 or more months of follow-up. Among these patients, 20 were premenopausal and one was postmenopausal at the time of recognition of the mammographic change (age range, 35–62 years; mean age, 44.2 years). In addition to mammography, eight patients underwent US, three underwent contrast material–enhanced magnetic resonance (MR) imaging, and two underwent both US and MR imaging. To identify any unifying characteristics, the histopathologic findings of 13 biopsies were reviewed by a pathologist (J.P.P.) who has particular interest in breast disease.
Among the 21 patients, 13 were asymptomatic and presented for screening examinations, and eight presented for diagnostic examinations. Of the diagnostic studies, seven were in asymptomatic patients who were referred from outside screening centers for evaluation of the enlarging asymmetry, and one patient presented for evaluation of focal pain that corresponded to the region of asymmetry.
Five of 21 patients had a close family history of breast cancer, including two patients younger than 40 years who had undergone annual or biennial screening mammograms for several years. One patient had an invasive ductal carcinoma in the contralateral breast that was diagnosed concomitantly with recognition of the asymmetry. All of the patients were parous (gravida 1–6, para 1–4). No patients were taking or had taken hormonal replacement medication, and none had taken birth control pills in the 5 years before recognition of the mammographic change. One patient had a cardiac history and was taking warfarin sodium. No patient had a history of biopsy or any other form of major trauma to the affected breast.
The mammograms were obtained with a CGR 500 T (GE Medical Systems, Milwaukee, Wis), a model MIII (LoRad, Danbury, Conn), or a Contour (Bennett Trex Medical, Copiague, NY) mammography unit. All patients underwent routine mammography, which consisted of craniocaudal and mediolateral oblique views. The mammographic work-up of the asymmetric tissue included at least one spot compression view in all cases in addition to a variety of other views, including mediolateral, exaggerated craniocaudal, and magnification views.
The US examinations were performed by using an Aloka 260 (Corometrics Medical Systems, Wallingford, Conn) with a 7.5-MHz linear-array transducer or an HDI Ultramark 9 (Advanced Technology Laboratories, Bothell, Wash) with a variable-frequency 10–5-MHz linear-array transducer. MR imaging examinations were performed with a Signa 1.5-T unit (GE Medical Systems) by using a dedicated dual breast coil (Medical Advances, Milwaukee, Wis). Sagittal MR imaging of the affected breast included a dynamically acquired three-dimensional, spoiled gradient-echo image acquired before and three times after the administration of 0.1 mmol of gadopentate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) per kilogram of body weight by using the following parameters: 17.1/4.2 (repetition time msec/echo time msec), 45° flip angle, and 1.6-mm section thickness. A delayed fat-suppressed, three-dimensional spoiled gradient-echo image (26.9/3.1) also was obtained.
Mammographically guided needle localization was performed for the 12 surgical biopsies. Three of the four core biopsies were performed by using stereotactic guidance; one of these procedures was reportedly performed by using an add-on device at an outside institution. The two other stereotactic procedures were performed by using a dedicated stereotactic table (Mammotest/Mammovision; Fischer Imaging, Denver, Colo). A 14-gauge, long-throw (2.3-cm excursion) needle and gun combination (Biopty; Bard Urological, Covington, Ga) were used for the stereotactic procedures performed at our institution and for a single US-guided biopsy. US guidance was used in one patient because the tissue in question was close to the nipple and difficult to stabilize in the stereotactic compression device. The tissue was confidently identified sonographically because with the exception of the enlarging focus of asymmetric tissue, the breast tissue was fatty.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The mean age of the remaining nine patients (ie, those with >0.5 cm of asymmetric tissue at baseline) was 41.6 years (range, 36–60 years) at baseline mammography and 46.9 years (range, 42–62 years) at the time of recognition of the mammographic change. The mean measurement of the asymmetric tissue was 2.4 cm (range, 1.0–4.0 cm) at baseline mammography and 4.6 cm (range, 2.0–5.5 cm) at the time of recognition of the mammographic change. Of eight patients who had enlargement of the asymmetric tissue, the mean change in size between that at baseline mammography and that at recognition of the change was 2.5 cm (range, 1.0–3.7 cm). One patient had asymmetric tissue at baseline mammography that increased markedly in radiographic density during 4 years but did not change appreciably in greatest dimension (4.0 cm) during 10 years.
Figure 1 shows the age distribution and rates of asymmetric tissue enlargement among the 20 patients who had a change in asymmetric tissue size. Among these patients, the mean increase in size from that at baseline mammography to that at biopsy or at follow-up mammography was 2.7 cm and ranged from 1.0 cm during 2 years to 6.5 cm during 5 years. There was a fluctuation in the rate of change (in absolute terms of centimeter per year) among individual patients. The greatest change in size between mammograms was calculated for each patient (mean, 0.8 cm/y; range, 0.3–2.0 cm/y). There were several patients in whom the asymmetric tissue, after increasing in size, showed a negligible change or a decrease in size during 1–3 years. There was no correlation between rate of change and patient age or presence of asymmetric tissue at baseline mammography.
|
Nine patients were examined with US. In four cases, no abnormality was identified, and only normal-appearing tissue was visualized. In the remaining five cases, the tissue in question was confidently identified and appeared heterogeneous but predominantly echogenic relative to the surrounding fatty tissue (Fig 2). In three cases, there were linear hypoechoic structures that were suggestive of ducts running within the more generalized area of hyperechogenicity.
|
|
|
|
|
|
|
|
|
|
In three cases, biopsy had been suggested by the radiologist (despite a negative US result in one case), but a follow-up course was taken instead. In one of these cases, because the US finding of hyperechoic tissue suggested that the asymmetry was benign, the surgeon opted for follow-up because the patient was a poor surgical candidate. The results of mammographic follow-up at 12 months showed no appreciable change. In another case, the patient's surgeon preferred follow-up over biopsy because the patient was asymptomatic and had normal physical examination results and the asymmetry had changed only slightly since a mammogram was obtained 2 years previously, although it had enlarged substantially from its size at baseline mammography.
One patient refused to undergo biopsy and subsequently has been followed up for 7 years (Fig 4). The asymmetric tissue measured 2.5 cm when it was first noted as a change in an area of the breast where no density had been present 2 years previously. The asymmetry enlarged gradually during 4 years to a maximum measurement of 6.5 cm. Between the 4th and 6th follow-up year, the patient went through menopause, and on her last two yearly mammograms, the asymmetry had decreased in density and size similarly to the regression of her parenchymal breast tissue.
|
|
|
|
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This phenomenon occurred predominantly in premenopausal women in their 5th decade of life. It is notable that at recognition of the mammographic change, the age of the 12 patients who had no asymmetric tissue at the baseline study (mean age, 41.7 years) was approximately the same as that of the nine women with baseline asymmetric tissue at the time of their first mammogram (mean age, 41.6 years). Similarly, the size of the asymmetric tissue at recognition of the change in the 12 patients without baseline asymmetric tissue (mean size, 2.4 cm) was similar to the size of the existing asymmetric tissue in the remaining women with baseline asymmetric tissue (mean size, 2.5 cm). This observation suggests that benign asymmetric tissue noted at baseline mammography in women around their 5th decade of life may have developed in the preceding decade.
In our group of 21 patients, only one (5%) was notably different in age and hormonal status from the others. She was postmenopausal and lacked a history of taking exogenous hormones or other medications. Note that the patients included in this review were recruited over time on a case-by-case basis and were not identified by reviewing all of the patients who presented to our breast imaging department. Therefore, the demographics of the population affected and the true prevalence of the phenomenon cannot be definitively described on the basis of our data set. In addition, our selection bias precluded the assessment of the outcome of all the patients with asymmetric tissue. Therefore, the likelihood of a coexistent malignancy in any patient with enlarging asymmetric tissue density is unknown.
The histopathologic findings in nearly all of our patients included prominent fibrosis. Haagenson (6) described a rare clinical entity, termed fibrous disease, that was found predominantly in premenopausal patients, manifested clinically as a palpable mass, and had distinctive microscopic features of perilobular fibrosis with obliteration of the acini. Sixty-two percent of the lesions developed in the upper outer portion of the breast, and there were three recurrences after excision; however, there was no increased prevalence of breast carcinoma in the 156 patients described (6). Note that five (3%) of the 156 patients were postmenopausal at the time of diagnosis, and two of these five women had taken hormones. Hermann and Schwartz (7) reported a single case of a 63-year-old patient (hormonal status undefined) who had a clinically occult 3 x 2-cm mass that at mammography had irregular borders at its posterior aspect and a rounded, more smoothly defined anterior margin. This mass was considered to be a mammographically detected case of the fibrous disease described by Haagenson (6,7).
The stromal change in our patients' specimens was predominantly pseudoangiomatous stromal hyperplasia. At microscopic examination, pseudoangiomatous stromal hyperplasia is recognized by slitlike, anastomosing spaces incompletely outlined by stromal cells, an appearance which closely resembles that of vascular spaces. The spaces have been attributed to the separation of collagen fibers (8). It is noted that this entity was rarely mentioned as a finding by the various pathologists who initially interpreted the biopsy results, although "stromal fibrosis" was frequently reported. This suggests that pseudoangiomatous stromal hyperplasia is generally not recognized as a distinct breast abnormality among pathologists.
Pseudoangiomatous stromal hyperplasia has been reported as an incidental microscopic finding of breast biopsy performed for either benign or malignant disease (23% of 200 consecutive breast biopsies) (9). It has also been reported, however, as the major histopathologic entity found in some discrete, usually palpable benign mass lesions in the breast (8,10–12). Hormonal factors are likely to contribute to pseudoangiomatous stromal hyperplasia (9–12). The average age of patients with this entity has been reported to be 40 years (9). To our knowledge, there is no evidence in the literature to suggest that pseudoangiomatous stromal hyperplasia is a premalignant entity or high-risk marker for malignancy, although there may be recurrence after excisional biopsy, and mastectomy has been reported for control of multiple nodular recurrences (10).
To our knowledge, little has been written with regard to the radiographic appearance of pseudoangiomatous stromal hyperplasia. Polger et al (12) reviewed seven cases in which this condition manifested as masses, three of which were palpable. Mammographically, the masses had circumscribed, indistinct, or obscured margins. US performed in five of these cases demonstrated discrete masses in four cases; in three of these cases, the masses were slightly heterogeneous but predominantly hypoechoic. The fourth mass that was imaged with US was heterogeneous and had a small cystic component. The growth of six of the masses was demonstrated either mammographically or clinically, and two patients had documented recurrences.
It would appear that the majority of our patients who had asymmetric breast tissue and underwent biopsy represent a subset of pseudoangiomatous stromal hyperplasia cases in which the entity is nonpalpable. Although we did not include patients who were receiving hormonal therapy in this report, the literature (11) suggests that the entity could occur in postmenopausal patients who are taking hormones. Whether the same histopathologic features account for the asymmetries in the patients who were followed up in our study is obviously unknown, although the similarities in morphology and behavior would suggest similar histopathologic findings.
In treating patients with increasing mammographic asymmetry, we emphasize that the work-up must focus on differentiating benign asymmetric breast tissue from more ominous focal asymmetric density. In cases of developing asymmetric breast tissue, negative results of a thorough work-up beyond mammography, including physical examination, US, and/or MR imaging, may suggest benign changes, and follow-up is appropriate. Asymmetric breast tissue may be caused by histopathologic findings, including fibrocystic changes, dense stromal fibrosis, and/or pseudoangiomatous stromal hyperplasia, and in the correct clinical and imaging settings, these findings may be considered to be diagnostic.
As it is our custom to follow other breast abnormalities that are found to be benign at core biopsy, we also recommend that these asymmetries be followed if core biopsy is performed to assess for changes that may suggest a missed malignancy in tissue that is not sampled. Conversely, if a benign biopsy specimen is obtained for a developing asymmetry, and the imaging and clinical features are benign as well, a repeat biopsy would not be efficacious and may cause unnecessary deformity if a large amount of tissue is removed. It should be noted that the focus may enlarge gradually if it is followed or partially removed, or it may recur following excisional biopsy. This phenomenon should be anticipated and explained to the patient. Continued management must be based on a combination of suspicious clinical and imaging findings.
|
Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, C.W.P.; study concepts and design, C.W.P., S.A.F.; definition of intellectual content, C.W.P., S.A.F.; literature research, C.W.P., S.A.F., J.P.P.; clinical studies, C.W.P., J.P.P.; data acquisition, C.W.P., J.P.P.; data and statistical analyses, C.W.P., S.A.F.; manuscript preparation, editing, and review, C.W.P., S.A.F., J.P.P.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. W. T. Leung and E. A. Sickles Developing Asymmetry Identified on Mammography: Correlation with Imaging Outcome and Pathologic Findings Am. J. Roentgenol., March 1, 2007; 188(3): 667 - 675. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
= patients who underwent open excisional biopsy, x = patients who underwent core biopsy, 
= patients who were followed up mammographically. No measurements were taken after core or excisional biopsy was performed.
