|
|
||||||||
Breast Imaging |
1 From the Departments of Radiology (S.C.H., C.M.D., P.J.D., D.N.S., J.E.M.) and Pathology (S.C.L.), Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115. Received March 25, 1998; revision requested June 15; revision received July 20; accepted October 19. Address reprint requests to S.C.H., Department of Radiology, University of Vermont College of Medicine, Fletcher Allen Health Care, University Health Center Campus, 1 S Prospect St, Burlington, VT 05401.
| Abstract |
|---|
|
|
|---|
MATERIALS AND METHODS: Of 853 breast masses in which LCNB was performed, 38 (4%) revealed histologic findings of fibrous nodules. Repeat biopsy was performed in 16 lesions (surgical excision, 13 lesions; repeat LCNB, three lesions). The mammographic and ultrasonographic findings in these 16 cases, the histologic LCNB findings, and the repeat biopsy findings were retrospectively reviewed.
RESULTS: Repeat biopsy findings confirmed the diagnosis of fibrous nodules in 15 of the 16 masses. In one mass, repeat LCNB findings showed a fibroadenoma. The imaging features of the fibrous nodules varied. Five nodules (33%) manifested as masses with indistinct margins; six (40%), as circumscribed masses. In four cases (27%), imaging suggested malignancy. None were calcified masses. An additional 22 masses had circumscribed or indistinct borders and were diagnosed with LCNB as fibrous nodules. None of these masses had suspicious findings, and repeat biopsy was not performed.
CONCLUSION: Fibrous nodules usually manifest as masses with circumscribed or indistinct margins but can have findings suggestive of malignancy. LCNB can reveal histologic findings consistent with this diagnosis. In all masses in which repeat biopsy was performed, the diagnosis of a fibrous nodule was confirmed or another benign stromal lesion was diagnosed.
Index terms: Breast, abnormalities, 00.31 Breast, biopsy, 00.1261, 00.1267, 00.12985 Breast, fibrous nodule, 00.3199 Breast neoplasms, US, 00.1298
| Introduction |
|---|
|
|
|---|
Percutaneous large-core needle biopsy (LCNB) has become widely used as an accurate and less invasive alternative to surgical biopsy (79). LCNB has a reported repeat biopsy rate of 14%18% (10,11) due to pathologic diagnoses of atypia, discordant histologic and imaging findings, and nonspecific histologic diagnoses. LCNB findings consistent with a fibrous nodule may result in repeat biopsy because the histologic findings are relatively nonspecific in isolation and, to our knowledge, the imaging features are not well described in the literature.
We performed this study to describe the imaging features associated with fibrous nodules and to describe the histologic features of LCNB specimens of this entity. In our series of breast masses sampled with LCNB, we encountered 38 cases in which the histologic findings were consistent with fibrous nodules. Repeat biopsy was performed in 16 of the lesions.
| MATERIALS AND METHODS |
|---|
|
|
|---|
In 807 masses (95%), LCNB was performed with 14-gauge long-throw (23-mm excursion) needle and gun combinations (Biopty or Monopty; Bard, Covington, Ga). In 391 of the 807 cases, LCNB was performed stereotactically by using a 14-gauge long-throw needle. In 416 of the 807 cases, LCNB was performed with US guidance by using a 14-gauge long-throw needle. A 14-gauge directional vacuum-assisted biopsy device (Mammotome; Biopsys Medical, Irvine, Calif) and stereotactic guidance were used in 46 (5%) of the LCNBs of masses. A mean of five core samples were obtained (range, three to seven) with the spring-loaded biopsy guns, and a mean of seven samples (range, two to 15) were obtained with the vacuum-assisted biopsy device.
To our knowledge, there are no published histologic criteria for the diagnosis of fibrous nodules in LCNB specimens. The specialized breast pathologist (S.C.L.) reviewed the cases where findings of both the initial LCNB and the subsequent excision showed dense collagenous stroma. The following histologic findings were characteristic of these lesions and were used as the criteria for the diagnosis of a fibrous nodule in subsequent cases. In each case, multiple core biopsy specimens showed collagenous stroma, with absent or only minute foci of adipose tissue (less than 10 adipocytes). The extent of the involved tissue measured at least 0.4 cm in length (range, 0.40.8 cm; mean, 0.65 cm). The epithelium usually appeared atrophic with thickened basement membranes. Lobules were either absent or lacked specialized intralobular stroma. Histologic features of other stromal lesions, including pseudoangiomatous stromal hyperplasia, fibroadenoma, lymphocytic or diabetic mastopathy, fibromatosis, nodular fasciitis, or myofibroblastoma, were absent (12).
Early in the study period, repeat biopsy was recommended due to the nonspecific histologic findings. In the early study period, repeat biopsy was performed in 12 (75%) of the 16 total lesions in which repeat biopsy was performed. Later in the study period, repeat biopsy was recommended only if the radiologic findings were unusual (eg, spiculated or irregular margins that raised the question of discordance) or if there was a concern about inadequate sampling; this happened in four (25%) of the 16 lesions.
The mammographic and US images were retrospectively reviewed together and at the same time by two experienced radiologists (S.C.H., C.M.D.). Mammographic images were evaluated for size, shape, and margins of the lesions. In addition, the US images were evaluated for echogenicity, heteroechogenicity, and posterior acoustic sound change. The histologic results from the initial LCNB and repeat biopsy specimens were retrospectively reviewed by an experienced subspecialized breast pathologist (S.C.L.) who was blinded to the radiologists' impression. The pathologist knew the results of the initial LCNB biopsy when reviewing the repeat biopsy slides.
| RESULTS |
|---|
|
|
|---|
Repeat biopsy was performed in 16 of the 38 lesions with the diagnosis of fibrous nodule. In 15 (94%) of the 16 lesions identified in a total of 14 women, findings of repeat tissue sampling confirmed the histologic diagnosis of a fibrous nodule. In the 16th lesion, repeat LCNB yielded fibroadenoma. A circumscribed 11-mm mass was shown at mammography that was circumscribed and solid at US.
The 14 women with a total of fibrous nodules confirmed with repeat biopsy were aged 3669 years (mean age, 45 years). Of these 14 women, 10 (71%) were premenopausal and four (29%) were postmenopausal. Three (75%) of the postmenopausal women were receiving hormone replacement therapy. None was known to be diabetic.
The lesions were 0.63.5 cm (mean, 1.8 cm). Twelve women had 13 nonpalpable mammographic masses (size range, 1.03.5 cm) at presentation. Two women had unilateral palpable masses (size range, 0.61.1 cm) at presentation. One woman had synchronous bilateral nonpalpable masses.
Of the 13 lesions manifesting as nonpalpable mammographic masses, 11 were round or oval. Six of the 11 had circumscribed margins (Fig 1a), five of the 11 had indistinct margins (Fig 2), and two were irregular in shape with spiculated margins (Fig 3a). None of the masses contained calcifications. Of the 13 nonpalpable masses, three were detected on baseline mammograms, two were larger than on prior mammograms, three were not seen on previous mammograms, and six were in women referred to our institution for biopsy without prior comparison images. US also was performed in 10 of these 13 masses; all were solid. Five of the six lesions circumscribed on mammograms also were circumscribed and homogeneously hypoechoic on sonograms (Fig 4) and were solid. Of the four of the five lesions with indistinct margins on mammograms, one homogeneously hypoechoic mass had indistinct margins and another homogeneously hypoechoic mass had irregular margins with posterior acoustic shadowing. The remaining two of the four indistinct lesions on mammograms were not visible at US. One of the two lesions with spiculated margins was imaged with US but was not visible.
|
|
|
|
|
|
|
Histologically, all the lesions consisted of dense collagenous connective tissue similar to that in adjacent breast tissue. Adipose tissue was absent or was present in only minute areas. In some lesions, a distinct border could be seen (Fig 1b). In most, the edge of the mass blended imperceptibly into adjacent stroma. Epithelial elements were scant and usually appeared atrophic. Diagnostic features of other stromal lesions (12,13), including pseudoangiomatous stromal hyperplasia, fibroadenoma, lymphocytic or diabetic mastopathy, nodular fasciitis, fibromatosis, or myofibroblastoma, were absent. In two lesions, however, there were focal areas suggestive of either fibroadenoma or pseudoangiomatous stromal hyperplasia.
The two spiculated lesions were unusual in that epithelial elements were absent in the center of the lesions (Fig 3b). The dense collagenous stroma resembled scar tissue and abutted adipose tissue, creating an irregular spiculated border. The lesions did not infiltrate the adjacent tissue. One woman did not have a history of breast surgery. The other woman had undergone reduction mammoplasty the previous year, but the lesion was not in the surgical site.
Histologic findings were consistent with fibrous nodule in 22 lesions sampled with LCNB in 22 women, but repeat biopsy was not performed in the lesions. These 22 women were aged 2475 years (mean age, 48 years). Of the seven women older than 50 years, four were receiving hormone replacement therapy. One was a postmenopausal woman not receiving hormone replacement therapy who developed the mass after starting tamoxifen citrate therapy for an ipsilateral carcinoma. Two were postmenopausal and were not receiving hormones.
Of these 22 masses, 17 (77%) were clinically occult and five (23%) were palpable. The size was 0.63.0 cm (mean, 1.5 cm). At mammography, eight of the nonpalpable masses were circumscribed, and six had indistinct margins. One nonpalpable mass was imaged only with US and was circumscribed. US was performed in an additional 11 of the nonpalpable masses. Eight of these were circumscribed masses, and three were not visible at US. All five palpable masses were evaluated with US; four were circumscribed, and one had indistinct margins. Two of the women with these masses also had mammographic studies. One mass was not visible due to dense parenchyma, and the other had circumscribed margins. None of the masses was US or mammographically highly suggestive of malignancy.
Repeat biopsy was recommended in two (9%) of the 22 masses that were followed up. Biopsy was recommended due to concern of sampling error; however, biopsy was refused in both cases. One of these two women underwent imaging 6 months later, and the mass was no longer visible; the other woman was lost to follow-up.
Follow-up was recommended in the remaining 20 lesions in 20 women in whom sampling with LCNB was considered adequate. Two lesions were stable or smaller at the 1-year follow-up examination, and six were stable or smaller at the initial 6-month follow-up examination. At the time the study was concluded, nine cases were current and were not yet due for follow-up. Three women with three lesions have been lost to follow-up.
| DISCUSSION |
|---|
|
|
|---|
Fibrous nodules were initially described as manifesting as firm, palpable masses in premenopausal women with histologic findings of collagenous stroma surrounding atrophic epithelial elements (1). Results in subsequent series confirmed these findings (25). Most lesions were unilateral, with bilateral masses present in 15%20%. Although the palpable masses were sometimes circumscribed at physical examination, characteristically the edge of the mass merges into surrounding dense tissue. After surgical excision, recurrence has been reported in less than 5% (1). To our knowledge, previous to this study, there was only one detailed case report of a fibrous nodule manifesting as a nonpalpable mammographic mass (6).
In this study, findings consistent with fibrous nodules were found in 4% of masses evaluated with LCNB. The frequency of fibrous nodules has previously been given as 3% of women with benign breast disease presenting to a clinic (3) or 7.9% of surgical specimens from the breast (5). Thus, although infrequent, fibrous nodules are not rare.
The pathogenesis of fibrous nodules is unknown. Their almost exclusive occurrence in premenopausal women 2050 years of age in previous series (1,3,5) as well as in our series suggests a hormonal role. Only 8% of the women in our series were postmenopausal and were not receiving hormone replacement therapy. Although the appearance has been compared with involutional changes in breast tissue (14), this is unlikely, given that the most common manifestation is a new nodule in a young woman.
The mammographic findings of a nonpalpable fibrous nodule were to our knowledge first described in a case report by Hermann and Schwarts (6), who reported a single, nonpalpable, 3 x 2-cm mass with partially obscured and partially irregular margins. To our knowledge, the range of mammographic and US appearances of masses with the confirmed histologic diagnosis of a fibrous nodule has not been reported. In our series, among the masses in which repeat biopsy was performed, most manifested with margins that were circumscribed (40%) or indistinct (33%) at mammography, US, or both. However, there were four masses (27%) with features highly suggestive of malignancy. Two of these were spiculated masses mammographically, one had indistinct margins at mammography and was irregular in shape with posterior shadowing at US, and one was palpable with an irregular shape and posterior shadowing at US. If all 38 masses in our series diagnosed as fibrous nodules by using initial LCNB are included (those in which repeat biopsy was performed and those in which it was not), 21 (55%) were circumscribed, 12 (32%) were indistinct, four (11%) were suggestive of malignancy, and one (3%) was diagnosed as a fibroadenoma after repeat biopsy.
The two cases in our series manifesting as spiculated masses were unusual radiographically and histologically. In contrast with the other lesions, which consisted of both stroma and sparse epithelial elements, these two lesions were almost exclusively collagenous stroma and resembled scar tissue. Diagnostic features of more typically spiculated stromal masses such as fibromatosis or nodular fasciitis were absent. One woman had no history of trauma or surgery, and the other had undergone reduction mammoplasty, but the surgery did not involve this area of the breast.
Successful core biopsy depends on optimizing technical factors and also on careful correlation of the imaging and the histologic results (7,10,1517). Therefore, familiarity with the imaging features of a fibrous nodule and recognition that it represents a discrete entity are important for appropriate management and follow-up in patients who undergo LCNB. This has been emphasized by Berg et al (18), who accept focal fibrosis as a specific benign histologic diagnosis for a discrete mass at mammography or US, provided the lesion does not appear highly suspicious. Berg et al (18) did not describe the imaging or histologic findings of focal fibrosis in detail, and it is important to be aware that descriptive pathologic evaluations may not correlate with the histologic findings of a fibrous nodule as defined in this study and could instead represent normal nondiagnostic breast tissue. As for LCNB in general, good communication between the radiologist and the pathologist is necessary for optimal correlation of results in cases of fibrous nodules. Lesions with features highly suggestive of malignancy, such as spiculated borders or posterior acoustic shadowing, although sometimes associated with fibrous nodules, should be evaluated after excision. A calcified lesion with the pathologic result of a fibrous nodule should be reevaluated, since none of the fibrous nodules in this series or others was calcified. In addition, a fibrous nodule would be unusual in a postmenopausal woman in the absence of hormone replacement therapy.
Although our study population is small, no breast cancers were missed and no atypia was found at repeat biopsy. In 15 (94%) of 16 cases, repeat biopsy confirmed the diagnosis of a fibrous nodule. In the remaining case, repeat LCNB findings were more compatible with a fibroadenoma. Thus, typical mammographic, US, or both mammographic and US findings combined with histologic findings consistent with a fibrous nodule can be used to place lesions into this specific benign category, and these nodules do not require further tissue sampling. These data also suggest that imaging follow-up of a fibrous nodule with low suspicion of malignancy at imaging is a safe strategy. However, additional tissue sampling should be performed in lesions with highly suspicious radiologic findings or if LCNB reveals nondiagnostic findings (10).
| Acknowledgments |
|---|
| Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, S.C.H.; study concepts, S.C.H., C.M.D., S.C.L.; definition of intellectual content, S.C.H., C.M.D., P.J.D., S.C.L.; clinical studies, P.J.D., D.N.S., J.E.M.; manuscript preparation, S.C.H., C.M.D., S.C.L.; manuscript editing, P.J.D., D.N.S., J.E.M.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. K. You, E.-K. Kim, J. Y. Kwak, M. J. Kim, K. K. Oh, B.-W. Park, and W. I. Yang Focal Fibrosis of the Breast Diagnosed by a Sonographically Guided Core Biopsy of Nonpalpable Lesions: Imaging Findings and Clinical Relevance J. Ultrasound Med., October 1, 2005; 24(10): 1377 - 1384. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Sklair-Levy, T. H. Samuels, C. Catzavelos, P. Hamilton, and R. Shumak Stromal Fibrosis of the Breast Am. J. Roentgenol., September 1, 2001; 177(3): 573 - 577. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Guenin, C. H. Lee, L. E. Philpotts, L. J. Horvath, and I. Tocino The Low False-Negative Rate for Stereotactic Breast Biopsy Dr Lee and colleagues respond: Radiology, August 1, 2000; 216(2): 609 - 610. [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |