| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Breast Imaging |
1 From the Department of Radiology (W.A.B., C.I.C.), Greenebaum Cancer Center (W.A.B.), and Department of Pathology (O.B.I.), University of Maryland Medical Center, 419 W Redwood St, Suite 110, Baltimore, MD 21201. From the 2001 RSNA scientific assembly. Received May 31, 2002; revision requested June 27; final revision received September 23; accepted September 30. Address correspondence to W.A.B. (e-mail: waberg@umaryland.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: From a database of 2,072 image-guided procedures performed from July 1995 through September 2001, 150 cystic lesions were identified. Diagnosis was established with fine-needle aspiration (n = 55), 14-gauge core-needle biopsy (n = 81), or both (n = 14). Excision was performed for all malignant (n = 18) and atypical (n = 2) lesions and for 11 benign lesions, which recurred or enlarged at follow-up. Imaging follow-up was available for 92 of 119 benign lesions. Targeted sonography was performed with high-frequency (10-MHz center frequency) transducers. Imaging and histopathologic, cytologic, and/or microbiologic findings were reviewed. Lesions were categorized as simple cysts, complicated cysts (imperceptible wall, acoustic enhancement, low-level echoes), clustered microcysts, cystic masses with a thick (perceptible) wall and/or thick (
0.5 mm) septations, intracystic or mixed cystic and solid masses (at least 50% cystic), or predominantly solid masses with eccentric cystic foci.
RESULTS: Of 150 lesions, 16 were simple cysts aspirated for symptomatic relief. Of 38 lesions characterized as complicated cysts and one cyst with thin septations, none proved malignant, nor did any of 16 lesions characterized as clustered microcysts. Of 23 masses with thick indistinct walls or thick septations, seven proved malignant. Of 18 intracystic or mixed cystic and solid masses, four proved malignant. Of 38 predominantly solid masses with eccentric cystic foci, seven proved malignant.
CONCLUSION: Symptomatic complicated cysts generally warrant aspiration. All clustered microcysts were benign, but further study is required. Cystic lesions with thick indistinct walls and/or thick septations (0.5 mm), intracystic masses, and predominantly solid masses with eccentric cystic foci should be examined at biopsy; 18 of 79 of such complex cystic lesions proved malignant in this series.
© RSNA, 2003
Index terms: Breast, biopsy, 00.1261, 00.1262 • Breast, cysts, 00.721, 00.722, 00.723 • Breast neoplasms, diagnosis, 00.311, 00.3115, 00.312, 00.319, 00.324, 00.327, 00.719 • Breast neoplasms, US, 00.12982, 00.12985, 00.12989
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Sonographic feature analysis of solid masses continues to improve (5), though observer variability continues to be problematic in avoiding biopsy (6,7). An illustrated Breast Imaging Reporting and Data System (BI-RADS) ultrasonographic lexicon (8) may be helpful in improving observer performance.
Confusion remains in the description and management of complicated cysts and complex cystic lesions of the breast (8–10). We sought to understand the pathologic basis of the sonographic features of cystic lesions of the breast and to determine appropriate assessment and management recommendations for these lesions based on sonographic appearance.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Our database is password protected, and our institutional review board did not require its approval or individual patient consent for review. Our database includes details of CNB and directional vacuum-assisted biopsies, aspirates when cytologic studies and/or cultures were performed, needle localizations, and ductograms. The database includes the mammographic and sonographic appearance of lesions as prospectively recorded. Of the 2,072 total procedures, 500 (24.1%) demonstrated a malignancy.
Follow-up after Fine-Needle Aspiration or CNB
Surgical excision was performed after any atypical (n = 2) or malignant (n = 18) result and was recommended for lesions with initial benign results, which recurred (n = 4) or increased (n = 7) at follow-up. The excised benign lesions were as follows: two recurrent cysts, two recurrent collections (one fat necrosis and/or seroma and one abscess), two papillomas that enlarged at follow-up, a third benign papilloma in a patient with persistent nipple discharge, one infarcted fibroadenoma 13 cm in diameter and another palpable fibroadenoma, one scar tissue developing at an abscess site, and one enlarging lactating adenoma.
If the lesion resolved after aspiration, and the cytologic specimen was benign, no further follow-up was recommended. Lesions proving to be benign cysts at CNB were not specifically followed up. If a proven abscess resolved clinically with antibiotics, no further follow-up was recommended. With a specific benign concordant result at CNB, such as fibroadenoma, lesions were followed up annually. For all other benign results, 6-, 12-, and 24-month follow-up was recommended. After a concordant benign result (other than cyst or abscess) at CNB or fine-needle aspiration, with no atypia, we accepted as a defined standard those lesions that were excised, subjected to another biopsy or reaspirated with the same benign result, gone or decreasing in size at any follow-up, or stable at 23 full months of follow-up. Of the 95 lesions subjected to CNB, 73 (77%) had an acceptable standard. Of the 55 lesions subjected to only aspiration, 53 (96%) had an acceptable standard.
Ninety-two proven benign lesions had at least one follow-up (range, 2–60 months; mean, 24 months; median, 23 months). Forty-nine lesions had at least 24 months of follow-up after initial diagnostic aspiration or CNB. Twenty-seven lesions proven benign at CNB (n = 16), aspiration (n = 8), or both (n = 3) have not had documented follow-up. The status of the 19 lesions after CNB without further follow-up include six fibrocystic changes; three fibroadenomas; two focal fibrosis; two abscesses; two papillomas; and one each sclerosing adenosis, apocrine metaplasia, cyst, and fat necrosis. The eight lesions with initial diagnosis at aspiration and no further follow-up include five simple cysts, one abscess, and two apocrine metaplasia (one sonographically simple cyst and one clustered microcysts, both in the same patient).
Imaging and Interpretation
Sonography was performed in all 150 lesions by using a broad-bandwidth linear-array transducer with a center frequency of 10 MHz supplemented by a transducer with a center frequency of 7.5 MHz, as needed, to penetrate larger breasts (Performa, Acoustic Imaging, Tempe, Ariz; or Elegra, Siemens Medical Systems, Issaquah, Wash). Craniocaudal and mediolateral oblique mammograms were obtained in 127 lesions by using a LoRad MIV unit (Hologic, Bedford, Mass) or DMR Plus unit (GE Medical Systems, Milwaukee, Wis). Spot compression views were obtained of all lesions manifested as asymmetric densities and of all palpable masses. Spot magnification views were obtained of all masses not clearly circumscribed on the original or spot compression views. Two radiologists (W.A.B. and C.I.C.) specializing in breast imaging reviewed the mammographic (n = 127) and sonographic (n = 150) images again and recorded features in consensus. All authors compared the imaging features with histopathologic and cytologic findings and/or culture diagnoses.
Anechoic masses with an imperceptible circumscribed border and acoustic enhancement were classified as simple cysts. Complicated cysts were defined by Mendelson et al (8) as lesions with homogeneous low-level echoes that otherwise meet the criteria of a simple cyst. Circumscribed cystic lesions with imperceptible wall and fluid-debris level were included as complicated cystic lesions. Lesions composed entirely of clusters of tiny (2–5 mm) anechoic foci with no discrete solid component were termed clustered microcysts. Septations within a cyst were defined as thin (<0.5 mm, representing the combined thickness of two myoepithelial and epithelial cell layers) or thick (0.5 mm). A thick wall was any perceptible wall.
Intracystic masses had a discrete solid mural mass within a cyst. Grouped with these were mixed cystic and solid masses with at least 50% cystic component. Masses that were at least 50% solid with eccentric cystic foci were considered the final category.
Sonographic and mammographic margins of the masses were recorded. For circumscribed lesions, the shape (round, oval, lobular, or tubular) was recorded. Lesions of mammographically focal increased density without convex borders were classified as focal asymmetries.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
0.5 mm) walls and/or thick (0.5 mm) septations, of which seven (30%) were malignant (Figs 5 –7). Of the seven malignancies, six (86%) were invasive ductal carcinomas (IDCs) grade 3, one of which had an associated intraductal component, and one was IDC grade 2 (Table 2). Of 18 abscesses, eight (44%) manifested as thick-walled cystic masses (Fig 8). Two abscesses with this appearance recurred: one was reaspirated and one was surgically incised and drained (Table 3). Another patient developed an irregular mass at the site of the initial abscess drainage; the mass was excised, showing scar tissue. Other benign causes of thick-walled cysts included simple cysts (including one with rupture and chronic inflammation), fibrocystic changes, and fat necrosis (Table 2). One fibrocystic change with atypia at cytologic examination recurred and was excised with benign results.
|
|
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The differentiation of a simple cyst from one with internal echoes or even solid lesions can be problematic at times, particularly if the lesion is deep or very small. With current transducers, most simple cysts 5 mm or larger can be reliably characterized with standard linear-array high-frequency transducers (center frequency of at least 10 MHz). Use of lower-frequency transducers, such as 7.5 MHz or lower, may be necessary for characterization of deeper cysts. Decrease in dynamic range can improve the ability to differentiate cystic and solid lesions. The use of tissue harmonic imaging may improve operator confidence in characterizing a lesion as a cyst, as has been shown in abdominal sonography (16), and is especially helpful in reducing artifactual internal echoes (17). Of 25 lesions in this series suspected to be complicated cysts that were initially aspirated, 22 (88%) resolved at aspiration and three (two abscesses and one fibrocystic change with apocrine metaplasia) had remaining thick walls. Another 13 (34%) of 38 lesions with the appearance of a complicated cyst in retrospect were subjected to initial diagnosis at CNB because they were believed to be solid at real-time evaluation.
Spatial compounding, in which the sonographic image is derived from the input of multiple off-perpendicular beams, may also facilitate evaluation of complicated cysts and masses that might be cysts or solid masses. In an evaluation of 166 cysts and/or complex cysts, spatial compounding increased diagnostic certainty in 77% of lesions (Stavros AT, personal communication, 2002). In 13 masses indeterminate for a cyst or a solid mass, diagnostic certainty was improved with spatial compounding in 85% of cases (Stavros AT, personal communication, 2002). Margin definition is similarly improved with spatial compounding (18). Spatial compounding is particularly helpful in decreasing speckle and other sources of noise; as a result, there is better definition of internal structure within masses (18). The improved signal-to-noise ratio also facilitates recognition of small cysts, though posterior enhancement is less apparent with spatial compounding (18). Merritt et al (18) found that the benefits of real-time spatial compounding decreased with increasing lesion depth.
In the series of Venta et al (10), only one (0.3%) of 308 complicated cysts proved malignant, which was a 3-mm focus of DCIS diagnosed at CNB. Buchberger et al (19) found none of 133 such lesions malignant, and Kolb et al (20) found none of 126 such lesions malignant in their series of screening sonography. In our series, none of 38 complicated cysts proved to be malignant. Thus, cumulatively, only one (0.2%) of 605 complicated cysts has proved to be malignant.
If a complicated cyst remains difficult to distinguish from a solid mass based on imaging characteristics alone, then we must consider the implications of following the mass if indeed it is solid. Of 2,072 breast imaging procedures in our database, three (0.1%) were believed to be complicated cysts, with aspiration initially attempted, but which proved entirely solid, yielding two fibroadenomas and one fibrocystic change at CNB. Each of these three solid masses were deep at the chest wall. Stavros et al (5) showed that nonpalpable, oval or gently lobulated, circumscribed, homogeneously hypoechoic masses with no distinctive posterior features and no suspicious features had less than 2% risk of malignancy, though this has not yet been reproduced at multiple centers. Possibly solid masses with posterior enhancement (as would be seen with complicated cysts) have not been so characterized.
Complex cystic masses with discrete solid components clearly require biopsy based on the results of this series. It is hoped, however, that through the use of spatial compounding, tissue harmonic imaging, decrease in the dynamic range, and increase in operator experience, lesions suspected to be complicated cysts will be followed up. Multicenter validation of this approach is planned. With further experience, we may find that when mobile internal echoes are depicted in such a lesion, classification as benign may be appropriate. Anecdotally, we have noted that complicated cysts are seen in patients with other simple cysts; an isolated complicated cyst versus a solid mass may not have the same risk of malignancy, though this concept also requires validation. A new or enlarging complicated cyst versus solid mass likely merits aspiration as the risk of malignancy may be greater in this context.
Symptomatic complicated cysts generally warrant aspiration. This would include tender or palpable complicated cysts. Again, as with simple cysts, we examine cytologic specimens only of bloody fluid. An abscess can appear as a well-defined cystic lesion with low-level internal echoes, and abscesses accounted for six (16%) of 38 complicated cysts in our series. When an abscess is suspected clinically at imaging or at inspection of the fluid, we send the fluid for culture and Gram staining and initiate a course of antibiotics that is appropriate for skin organisms. In the appropriate clinical setting, hematomas, fat necrosis, and galactoceles can appear as complicated cysts. Management should be predicated on the basis of patient symptoms.
Of 35 lesions proving to be cysts, seven (20%) recurred after initial aspiration. Cysts with a proliferative apocrine lining have been suggested to be at higher risk of recurrence than those with flat epithelium (21) and to carry an increased risk of malignancy (22) that is on the order of two-fold relative risk (23,24). Indeed, Bodian et al (24) suggested that the risk of malignancy was proportional to the number of cyst aspirations (which in turn may be related to proliferative changes in the cysts), with women who had undergone more than 10 cyst aspirations at up to a six-fold increased risk of breast cancer. None of the recurrent cysts in our series had apocrine lining, and the two recurrent cysts that were excised proved to be benign. With the detail apparent with use of current sonographic transducers, it is doubtful that excision is necessary for a recurrent simple cyst unless it is desired by the patient. The instillation of air at the time of aspiration appears to decrease the risk of recurrence (25), but it did not prevent recurrence of a 7-cm postoperative seroma with fat necrosis in our series.
In an overlapping series, we previously suggested that pure clusters of microcysts without discrete solid components can be considered probably benign (26) and be followed up. Such lesions are often due to apocrine metaplasia, though fibrocystic changes without apocrine metaplasia can have a similar appearance. It is possible that larger cysts with thin septations represent the continuum of the spectrum from apocrine metaplasia to cysts as acini fuse (26). In our collective experience (27), we have biopsy proof (n = 18) or 2-year follow-up results (n = 48) of 66 lesions prospectively characterized as clustered microcysts with no malignancies, and we now follow such lesions at 6, 12 and 24 months. Multicenter validation of this approach is required and planned.
A thick wall or thick septations (0.5 mm) in an otherwise cystic lesion without antecedent trauma or evidence of infection should suggest possible malignancy and prompt biopsy. We found that seven (35%) of 20 lesions without such relevant history were malignant, with six (86%) of seven due to high-grade IDCs. With current transducers, the sonographic borders of thick-walled cystic lesions are more typically indistinct, though the mammographic appearance may be circumscribed. While low-grade IDC is more often spiculated, circumscribed borders can occur with high-grade malignancies, which were seen in 16% of such lesions in the series of Lamb et al (28). Of eight high-grade IDCs in our series, three (33%) were circumscribed at sonography. One of these high-grade cancers had medullary features, and the others were IDC not otherwise specified.
While aspirates can demonstrate malignant cells, the cystic component of thick-walled cysts can be necrotic or acellular (and thereby nondiagnostic), and CNB of the wall or septations is preferred to provide a specific diagnosis and facilitate definitive treatment. Indeed, the cystic component of high-grade malignancies in this series represented areas of necrosis. Placement of a clip or an embolization coil marker at the time of sonographic biopsy (29–31) will facilitate subsequent identification of the lesion if excision is required. This is particularly important for lesions that are small (
7 mm in our experience) or appear to resolve after aspiration or CNB or if neoadjuvant chemotherapy is planned (32).
Abscesses, apocrine metaplasia, and inflamed or ruptured cysts or ducts can also present as thick-walled cysts. Hematomas can also have thick walls or thick septations. Fat necrosis can manifest as a thick-walled cystic lesion or a mixed complex cystic and solid mass, accounting for seven (23%) of 31 lesions with fat necrosis in one series (33) and nine (8%) of 112 in another (34). Short follow-up (2–3 months until resolution) of a thick-walled cystic lesion may be appropriate in the proper clinical setting of trauma or signs of infection. Any enlargement of such a lesion should prompt biopsy, as an underlying malignancy can either bleed due to trauma or cause obstruction of ducts and predispose the patient to infection.
Both benign and malignant papillary lesions often manifest as intraductal or intracystic masses (35–37), and three (75%) of four malignant intracystic masses were papillary in this series. Intracystic papillary carcinoma remains uncommon, accounting for only 0.6% of breast cancer in one series (38) and for 0.8% (four of 500) of breast cancers in our biopsy population. Caution is appropriate in accepting benign papilloma as a diagnosis at CNB, because malignancy and atypia can be focal in a large papilloma. Mammographic- and sonographic-histopathologic concordance, including size concordance, are critical with papillary lesions, as benign papillomas can also be an incidental finding adjacent to cancers, with such malignancies reported in two (17%) of 12 lesions excised after a CNB diagnosis of benign papilloma in two series (39,40). Any irregular or indistinctly marginated masses yielding benign papilloma at CNB should be considered discordant and prompt excision (40). Clearly, atypical papillary lesions should be excised, with six (30%) of 20 proving malignant in several small series (39–41). In the review of Reynolds (42), one (50%) of two sclerosed papillomas proved to be malignant.
Eccentric cystic foci can be due to dilated ducts, acini, or necrosis. Fibroadenomas rarely can have eccentric cystic foci, as was first described by Jackson et al (43). Only 12 (2.9%) of 407 fibroadenomas in our database had cystic foci. The presence of cystic foci in a mass that is otherwise consistent with a fibroadenoma may suggest a possible phyllodes tumor, though such tumors are rare. Cystic foci were seen in seven (23%) of 30 phyllodes tumors in the series of Liberman et al (44) and were more common in malignant phyllodes tumors. Apocrine metaplasia can coexist with a variety of benign lesions, including fibroadenoma, and be evident as eccentric cystic foci (26). Malignancies with eccentric cystic foci had no particular distinguishing features in our series, with low- and high-grade IDC, invasive lobular carcinoma, and intraductal carcinoma rarely having this appearance.
The margin characteristics of cystic lesions also merit discussion. For nonpalpable mammographically circumscribed masses, Sickles (45,46) and Varas et al (47) have shown that the risk of malignancy is less than 2% and that such lesions can be followed up. Indeed, none of the 29 nonpalpable sonographically circumscribed round or oval lesions were malignant in our series. Of the 54 palpable circumscribed round or oval lesions, five (9%) proved to be malignant. Those five malignancies were seen as intracystic masses (n = 2), thick-walled cystic masses (n = 2), and a predominantly solid mass with eccentric cystic foci (n = 1). All sonographically circumscribed lobulated masses that were malignant in this series were complex cystic and solid masses. As has previously been suggested (48), sonography may be helpful in identifying those circumscribed masses that merit biopsy.
In conclusion, sonographically simple cysts can be dismissed as benign. On the basis of the collective experience of this series and of the literature, the majority of incidental nonpalpable complicated cysts containing low-level echoes or fluid-debris levels may be be classified as probably benign, though further study is warranted and planned. Symptomatic complicated cysts should be managed on clinical grounds and generally warrant aspiration, with abscess, hematoma, fat necrosis, and galactocele in the differential diagnosis. In this series, all clustered microcysts were benign, though further study is required. Complex cystic breast masses with thick walls and thick septations, intracystic masses, masses with mixed cystic and solid components, and solid masses with eccentric cystic foci merit biopsy, with 18 (23%) of 79 of such complex cystic lesions proving to be malignant in our series.
|
FOOTNOTES |
|---|
Abbreviations: CNB = core-needle biopsy, DCIS = ductal carcinoma in situ, IDC = invasive ductal carcinoma
Author contributions: Guarantor of integrity of entire study, W.A.B.; study concepts and design, W.A.B.; literature research, W.A.B., C.I.C.; clinical studies, W.A.B., C.I.C., O.B.I.; data acquisition and analysis/interpretation, W.A.B., C.I.C., O.B.I.; manuscript preparation and definition of intellectual content, W.A.B.; manuscript editing, revision/review, and final version approval, W.A.B., C.I.C., O.B.I.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. H. Shim, E. J. Son, E.-K. Kim, J. Y. Kwak, J. Jeong, and S. W. Hong Benign Intracystic Papilloma of the Male Breast J. Ultrasound Med., September 1, 2008; 27(9): 1397 - 1400. [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Doshi, D. E. March, G. M. Crisi, and B. F. Coughlin Complex Cystic Breast Masses: Diagnostic Approach and Imaging-Pathologic Correlation RadioGraphics, October 1, 2007; 27(suppl_1): S53 - S64. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. C. Booi, P. L. Carson, M. O'Donnell, M. S. Richards, and J. M. Rubin Diagnosing Cysts With Correlation Coefficient Images From 2-Dimensional Freehand Elastography J. Ultrasound Med., September 1, 2007; 26(9): 1201 - 1207. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Mesurolle, T. Helou, M. El-Khoury, M. Edwardes, E. J. Sutton, and E. Kao Tissue Harmonic Imaging, Frequency Compound Imaging, and Conventional Imaging: Use and Benefit in Breast Sonography J. Ultrasound Med., August 1, 2007; 26(8): 1041 - 1051. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Graf, T. H. Helbich, G. Hopf, C. Graf, and E. A. Sickles Probably Benign Breast Masses at US: Is Follow-up an Acceptable Alternative to Biopsy? Radiology, July 1, 2007; 244(1): 87 - 93. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-W. Chang, K. H. Kwon, D. E. Goo, D. L. Choi, H. K. Lee, and S. B. Yang Sonographic Differentiation of Benign and Malignant Cystic Lesions of the Breast J. Ultrasound Med., January 1, 2007; 26(1): 47 - 53. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. A. Berg, J. D. Blume, J. B. Cormack, and E. B. Mendelson Operator Dependence of Physician-performed Whole-Breast US: Lesion Detection and Characterization. Radiology, November 1, 2006; 241(2): 355 - 365. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Bernathova, B. Zelger, D. Deardon, and G. Bodner Sonographic findings of an intraductal mucinous carcinoma. J. Ultrasound Med., July 1, 2006; 25(7): 925 - 927. [Full Text] [PDF]
|
|
|
|
 |
|
 W. A. Berg Sonographically Depicted Breast Clustered Microcysts: Is Follow-Up Appropriate? Am. J. Roentgenol., October 1, 2005; 185(4): 952 - 959. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||
