Ultrasonographic hyperechoic lesions of the breast: are they always benign?

Abstract

Background

Most of the breast lesions show hypoechogenicity relative to fat on ultrasonography. The frequency and malignancy rate of hyperechoic lesions are not investigated in a large series.

Purpose

To evaluate the frequency and malignancy rate of hyperechoic lesions on breast sonography and to investigate sonographic characteristics that may predict malignancy in hyperechoic breast lesions.

Material and Methods

Radiologic reports of 16,416 patients who underwent breast sonography between 2007 and 2008 were searched using “hyperechoic”, “echogenic” or “heterogeneous echoic” to describe lesions. Sonographic findings were evaluated according to the Breast Imaging Reporting and Data System lexicon. Clinical records including follow-up and pathologic findings were also reviewed. We calculated the frequency of hyperechoic lesions and their malignancy rate. Differences in sonographic appearances between benign and malignant lesions were also investigated.

Results

Among the 16,416 patients, 103 (0.6%) hyperechoic lesions were identified (mean size, 1.79 cm). Of these 103 lesions, 27 (26.2%) were pathologically evaluated and five (4.9%, 4 invasive ductal carcinoma and 1 mucinous carcinoma) were confirmed as malignant. Among the 819 malignant lesions diagnosed using sonography-guided core needle biopsy, five (0.6%) were hyperechoic. In benign lesions, fat necrosis and fibroadenoma were common pathologic diagnoses. Malignant lesions were more likely to have irregular shape (P = 0.003), non-parallel orientation (P = 0.002), non-circumscribed margin (P = 0.007), and a hypoechoic area (P = 0.027) than benign lesions. All hyperechoic carcinomas were seen as suspicious masses on mammograms.

Conclusion

Hyperechoic masses are very rare and mostly benign. As an adjunct to mammography, the imaging findings reported here could help to avoid misdiagnosis for malignant hyperechoic lesion.

Keywords

Breast neoplasm hyperechogenicity sonography pathology

Introduction

With advances in the resolution of sonography, breast sonography has proved useful in the evaluation of masses detected using mammography or clinical examination. A Breast Imaging Reporting and Data System (BI-RADS) (1) lexicon for breast sonography is commonly used to describe breast masses and to differentiate benign from malignant masses (2 –4). This lexicon includes descriptors of features such as mass shape, margin, orientation, echogenicity, and posterior acoustic features. Most breast lesions show hypoechogenicity relative to fat on sonography, and hypoechogenicity is generally accepted as a predictive feature for malignancy (3 –5), whereas hyperechogenicity is predictive of benignancy (2,3,5,6). Although hyperechoic lesions are generally regarded as benign, their clinical significance is not well known.

According to previous reports, the rate of hyperechoic malignancy is in the range of 0.4–2% (4,7,8). In a recent study, Linda (7) showed that 0.6% of all biopsied lesions (25/4511) were hyperechoic and 0.5% of malignant lesions were hyperechoic (9/1849). However, this study included only biopsy-proven breast masses. To our knowledge, no studies have been conducted on whole breast sonography examinations to determine the prevalence, clinical significance, and radiologic findings predicting malignant hyperechoic lesions on breast sonography. Therefore, the purposes of our study were to evaluate the frequency and malignancy rate of hyperechoic lesions on breast sonography and to investigate sonographic characteristics that may predict malignancy.

Material and Methods

Patients

Institutional Review Board approval was obtained for this retrospective study and informed patient consent was waived.

We reviewed a database of all breast sonography examinations accounting for 16,416 patients of 26,872 examinations that were performed between January 2007 and December 2008. Data collected included sonography-guided core needle biopsy (n = 2370) and sonography-guided vacuum-assisted biopsy (n = 318) reports. A search for radiologic reports that included the words “hyperechoic”, “echogenic”, or “heterogeneous echoic”, identified 160 patients that presented with hyperechoic lesions on breast sonography. In our study, a hyperechoic lesion was defined as one predominantly composed of areas of increased echogenicity relative to fat or echogenicity equal to that of fibroglandular tissue in >70% of the lesions. We excluded 57 cases in which intracystic or intraductal lesions or lesions that appeared hyperechoic because of calcification were observed. Finally, 103 patients were included in this study.

Imaging and outcome analysis

We used HDI 5000 or iU22 (Philips Medical Systems, Bothell, WA, USA) ultrasound scanners equipped with a commercially available 12-5-MHz linear-array transducer. Breast sonography was performed by one of five board-certified radiologists or one of four senior residents. Core needle biopsy was performed under sonography guidance using a 14-gauge biopsy gun (Acecut, TSK Laboratory, Soja, Japan) and vacuum-assisted biopsy was performed using 11 - or 8-gauge vacuum probes (Mammotome; Devicor Medical, Cincinnati, OH, USA). Mammography was performed using digital mammographic equipment (Senographe 2000D; GE Healthcare, Buc, France). Breast magnetic resonance imaging (MRI) was performed using a 3-T system (Achieva; Philips Medical Systems, Best, The Netherlands) with a dedicated bilateral breast coil and the patient in the prone position. The sonograms were retrospectively evaluated according to the BI-RADS lexicon by two board-certified radiologists in consensus (1). The radiologists assessed each lesion for shape (oval, round, irregular), margins (circumscribed, not circumscribed; indistinct, angular, microlobulated, spiculated), orientation (parallel, non-parallel), presence or absence of hypoechoic areas (present, absent), posterior acoustic features (normal, enhancement, shadowing), and vascularity (increased, absent, not evaluated). The maximum diameter of each lesion was measured. Mammograms were retrospectively reviewed by two breast radiologists for focal asymmetry, masses, and architectural distortion, according to the BI-RADS lexicon. If a mass was present, the shape (oval, round, lobular, or irregular), margin (circumscribed or not circumscribed; microlobulated, obscured, indistinct, spiculated), and density (high density, isodense, or low density) were noted according to the BI-RADS lexicon. Any discrepancy in opinion was resolved through consensus. We retrospectively reviewed clinical records, follow-up results, and pathologic findings.

Statistical analysis

The frequency of hyperechoic carcinomas among all carcinomas diagnosed at sonographically-guided biopsy as well as the frequency of hyperechoic lesions among all breast sonography examinations was calculated. Differences in sonographic appearance between benign and malignant lesions were investigated using the chi-square test or the Fisher exact test. All statistical analyses were performed using SPSS software version 19.0 (SPSS Inc., Chicago, IL, USA), and a P value of <0.05 was considered to indicate a statistically significant difference.

Results

Among the 16,416 patients, including 2688 patients who underwent sonography-guided core needle biopsy or vacuum-assisted biopsy, 103 (0.6%) examinations in 103 women (age range, 27–78 years; mean age, 52 years) were classified as hyperechoic in the original report and after retrospective image analysis. The mean size of lesions was 1.8 cm (range, 0.5–15 cm). Among 103 lesions, 30 (29.1%) were palpable masses and 73 (70.9%) were non-palpable masses. Of the 103 lesions, 27 (26.2%) were pathologically confirmed: five (4.9%) were identified as malignant and 22 (21.4%) were identified as benign. Among 2688 lesions biopsied, 819 proved to be malignant. Of these 819 biopsy-proven carcinomas, five (0.6%) were hyperechoic. Hyperechoic lesions with malignant pathologic diagnoses included invasive ductal carcinoma (n = 4) (Figs. 1 –3) and mucinous carcinoma (n = 1). Clinical and imaging findings of these five hyperechoic malignancies are summarized in Table 1. Of the five patients with hyperechoic carcinomas, three were palpable masses. All lesions were classified as more than BI-RADS category 4 on sonography. All hyperechoic carcinomas were correctly detected on initial mammographic evaluation in five patients. Upon mammography, all patients presented with a mass. In the five cases with masses observed upon mammography, the shape of the mass was irregular in three (60%) patients and lobular in two (40%) patients. The margin of the mass was indistinct in three (60%) patients, microlobulated in one (20%) patient, and circumscribed in one (20%) patient. The density of the mass was high density in all (100%) patients. Three patients underwent partial mastectomy and two underwent total mastectomy after core needle biopsy. Biopsy-proven benign lesions included fat necrosis (n = 9) (Fig. 4), fibroadenoma (n = 6) (Fig. 5), lipoma (n = 2) (Fig. 6), hamartoma (n = 1), angiolipoma (n = 1), abscess (n = 1), adenosis (n = 1), and benign epithelial proliferative lesion (n = 1).

Fig. 1.

A 61-year-old woman with grade 3 invasive ductal carcinoma in the right breast. Sonogram showing a 28-mm predominantly hyperechoic mass (arrows) with irregular shape and a central hypoechoic area in the upper inner quadrant of the right breast.

Fig. 2.

A 45-year-old woman with invasive ductal carcinoma in right breast. Sonogram showing a 18-mm oval predominantly hyperechoic mass (arrows) with indistinct margins and a central hypoechoic area.

Fig. 3.

A 62-year-old woman with invasive ductal carcinoma in right breast. Sonogram showing a 8-mm irregular predominantly hyperechoic mass (arrows) with indistinct margins and non-parallel orientation.

Fig. 4.

A 73-year-old woman with a palpable lump in the upper outer quadrant of the right breast. Correlative sonogram showing multiple hyperechoic masses (arrows) with irregular shape and indistinct margins. Sonography-guided core needle biopsy revealed fat necrosis. Six-month sonographic follow-up showed interval regression.

Fig. 5.

A 56-year-old woman with fibroadenoma. (a) Mammogram showing an oval isodense mass (arrows) with circumscribed margins in the lower outer quadrant of the left breast. (b) Corresponding sonogram showing a 14-mm predominantly hyperechoic mass (arrows) with circumscribed margins. Twelve-month sonographic follow-up showed no interval changes.

Fig. 6.

A 50-year-old woman with lipoma. (a) Mammogram showing a 70-mm oval fat-containing mass (arrows) with circumscribed margins in the retromammary fat layer of the right upper breast. (b) Correlative sonogram showing an oval hyperechoic mass (arrows) with circumscribed margins and parallel orientation. Pathology result after surgical excision revealed lipoma.

Table 1.

Clinical and imaging findings of hyperechoic carcinomas.

No.	Palpable	Mammographic finding	Mammographic BI-RADS category	Size (mm)	Sonographic feature						BI-RADS category	Histologic results
No.	Palpable	Mammographic finding	Mammographic BI-RADS category	Size (mm)	Shape	Orientation	Margin	Hypo-echoic area	Posterior acoustic feature	Vascularity	BI-RADS category	Histologic results
1	No	Mass	4	5	Irregular	Non-parallel	Indistinct	Present	Normal	Not evaluated	5	IDC
2	No	Mass	4	8	Irregular	Non-parallel	Indistinct	Present	Normal	Not evaluated	4	IDC
3	Yes	Mass	4	32	Irregular	Parallel	Microlobulated	Present	Normal	Absent	4	IDC
4	Yes	Mass	4	18	Oval	Parallel	Circumscribed	Present	Normal	Increased	5	IDC
5	Yes	Mass	5	150	Oval	Parallel	Circumscribed	Present	Normal	Absent	4	Mucinous carcinoma

IDC, Invasive ductal carcinoma.

Table 2 shows the sonographic findings of 103 hyperechoic lesions of the breast. Analysis of the sonographic features of malignant lesions indicated that the most frequent findings were irregular shape (60.0%) (Fig. 1), non-circumscribed margins (60.0%) (Fig. 1), parallel orientation (60.0%), presence of a hypoechoic area (100.0%) (Fig. 1), normal posterior acoustic features (100.0%), and absence of vascularity (40.0%). With regard to benign lesions, the most frequent sonographic findings were oval shape (86.7%) (Fig. 5), circumscribed margin (90.8%) (Fig. 6), parallel orientation (100%), presence of a hypoechoic area (46.9%) (Fig. 4), normal posterior acoustic features (78.6%), and absence of vascularity (37.8%). Malignant lesions were more likely to have irregular shape (P = 0.003), non-parallel orientation (P = 0.002), non-circumscribed margin (P = 0.007), and presence of a hypoechoic area (P = 0.027) than benign lesions (Fig. 1).

Table 2.

Sonographic findings of 103 hyperechoic lesions of the breast.

Malignant lesions (n, %) (n = 5)	Benign lesions (n, %) (n = 98)	P value
Shape	0.003
Oval	2 (40.0)	85 (86.7)
Round	0 (0.0)	9 (9.2)
Irregular	3 (60.0)	4 (4.1)
Margin	0.007
Circumscribed	2 (40.0)	89 (90.8)
Not circumscribed	3 (60.0)	9 (9.2)
Indistinct	2 (40.0)	8 (8.2)
Angular	0 (0.0)	0 (0)
Microlobulated	1 (20.0)	1 (1.0)
Spiculated	0 (0)	0 (0)
Orientation	0.002
Parallel	3 (60.0)	98 (100.0)
Non-parallel	2 (40.0)	0 (0)
Hypoechoic area	0.027
Present	5 (100.0)	46 (46.9)
Absent	0 (0.0)	52 (53.1)
Posterior acoustic feature	1.000
Normal	5 (100.0)	77 (78.6)
Enhancement	0 (0.0)	6 (6.1)
Shadowing	0 (0.0)	15 (15.3)
Vascularity	0.377
Increased	1 (20.0)	5 (5.1)
Absent	2 (40.0)	37 (37.8)
Not evaluated	2 (40.0)	56 (57.1)

Among the 76 lesions that were not confirmed pathologically, 55 (72.4%) lesions were followed-up sonographically and 21 were not. The mean follow-up period was 36 months (range, 6–60 months; median, 36.5 months). Follow-up sonography showed stable (n = 25, 45.5%) or decreased (n = 30, 54.5%) lesion size in all cases.

Discussion

Although breast cancers are commonly seen as hypoechoic relative to the surrounding tissue, some solid lesions may present with mixed echogenicity or hyperechogenicity because of their histological components. Hyperechogenicity is defined as increased echogenicity relative to fat or echogenicity equal to that of fibroglandular tissue (3). Several authors have reported that hyperechoic lesions are always benign (2,3,5,6). However, hyperechoic malignancies have been reported at various rates ranging between 0.5% and 2%. Linda et al. (7) found nine (0. 5%) hyperechoic cancers among 1849 malignant lesions on sonography-guided core needle biopsy and Soon et al. (8) found two (0.5%) hyperechoic cancers among 393 screen-detected breast cancers. Skaane and Engedal (4) found that four (2%) of 194 invasive ductal carcinomas were hyperechoic. In this study, we found 103 hyperechoic lesions of the breast, corresponding to 0.6% of all breast sonography examinations. Of 103 hyperechoic lesions, five (4.9%) were malignant, accounting for 0.03% of all breast sonography examinations performed and 0.6% of all malignant lesions diagnosed by sonography-guided biopsy. The malignancy rate of hyperechoic breast lesions reported in our study (0.6%) was similar to that reported by Linda et al. (0.5%) (7) and Soon et al. (0.5%) (8).

Many different pathologic conditions can present as hyperechoic breast masses on sonography. Hyperechogenicity can result from the presence of closely packed small adipocytes or clustered fibrous streaks, the interface of fibrous and adipose tissue surrounding and insinuating itself between adenomatous portions, or the presence of multiple and small vascular channels in benign conditions (9 –11). Benign lesions of the breast that present as hyperechoic masses include fat necrosis (12), lipoma (7, 13), fibroadenoma (4), hamartoma (14), focal fibrosis (9), angiolipoma (15), and hemangioma (16). In our study, fat necrosis and fibroadenoma were the most frequent histopathologic types observed. Although hyperechoic malignancy of the breast is rare, previous studies have shown that hyperechoic and isoechoic patterns are more frequent in invasive lobular carcinomas than in tumors of other histology types (17,18). However, the most common type of hyperechoic malignancy in this study was invasive ductal carcinoma because of its higher incidence in the general population (3). Tumor cell heterogeneity may play an important role in producing hyperechogenicity in malignant conditions (19). Uncommon malignant lesions of the breast that may present as hyperechoic nodules include lymphoma (20), angiosarcoma (21), and metastases (22).

Analysis of sonographic features showed that hyperechoic malignancies demonstrate more frequently irregular shape, non-parallel orientation, and non-circumscribed margins than benign hyperechoic nodules with a statistical significance (P = 0.003, 0.002, 0.007, respectively). However, sonographic features such as parallel orientation and normal posterior acoustic features were reported among the most frequent findings of these malignant lesions, which is in contrast with the sonographic malignant features reported by Stavros et al. (3). This may be due to the extremely small number of malignant lesions and therefore, careful interpretation of our results is needed. Although admitting this limitation, our results suggest that, in cases of hyperechoic nodules, the same sonographic features used for characterization of hypoechoic or isoechoic nodules could be applied to distinguish malignant from benign hyperechoic masses (1 –3,5). Especially, similarly to Linda et al. (7), we found that the presence of focal hypoechoic areas within the hyperechoic lesion was significantly more frequent in malignant nodules (P = 0.027). None of the five malignancies in our study was a purely hyperechoic lesion. In contrast, nine (40.9%) biopsy-proven benign lesions were purely hyperechoic lesions. We anticipate these ancillary findings can be helpful in predicting the malignant nature of the lesion. There may be potential for criticism regarding the inclusion of heterogeneously hyperechoic lesion. The criteria that defined hyperechogenicity are not well established. On the contrary to other studies (7,8,17) and ours, Stavros et al. (3) found that 42 biopsied hyperechoic lesions proved to be benign, concluding that hyperechogenicity was a benign feature with a 100% negative predictive value for malignancy. However, in their study, only homogeneously hyperechoic lesions were included. The authors indicated that if areas of isoechogenicity or hypoechogenicity that were larger than normal ducts or terminal ductal-lobular units (>4 mm) existed within a hyperechoic lesion and were not entrapped fat lobules, the areas could represent small malignant nodules.

Our study had several limitations. First, because we retrospectively searched radiologic reports, cases in which hyperechogenicity was not initially reported were excluded. This may have resulted in under-representation of hyperechoic lesions. Second, assessment was limited to a few static images, which is not typical of routine practice where evaluation of sonographic features is performed in real time during the examination. Third, because of the rarity of hyperechoic breast lesions (103 in this study) and the small number of cases that were subsequently confirmed as malignant, it was difficult to calculate predictive values of each sonographic finding.

In conclusion, the frequency and malignancy rate of hyperechoic lesions on breast sonography were 0.6% and 0.03%, respectively. A hyperechoic mass is very rare and most often benign. Nonetheless, this feature is unable to solely exclude malignancy. Identification of sonographic features that can predict malignancy, as described in this study, may help avoid misdiagnosis when used as an adjunct to mammography.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

American College of Radiology. BI-RADS atlas: breast imaging reporting and data system atlas , 4th edn. Reston, VA: American College of Radiology, 2003.

Rahbar

Sie

Hansen

. Benign versus malignant solid breast masses: US differentiation. Radiology 1999; 213: 889–894.

Stavros

Thickman

Rapp

. Solid breast nodules: use of sonography to distinguish between benign and malignant lesions. Radiology 1995; 196: 123–134.

Skaane

Engedal

. Analysis of sonographic features in the differentiation of fibroadenoma and invasive ductal carcinoma. Am J Roentgenol 1998; 170: 109–114.

Hong

Rosen

Soo

. BI-RADS for sonography: positive and negative predictive values of sonographic features. Am J Roentgenol 2005; 184: 1260–1265.

Del Frate

Bestagno

Cerniato

. Sonographic criteria for differentiation of benign and malignant solid breast lesions: size is of value. Radiol Med 2006; 111: 783–796.

Linda

Zuiani

Lorenzon

. Hyperechoic lesions of the breast: not always benign. Am J Roentgenol 2011; 196: 1219–1224.

Soon

Vallentine

Palmer

. Echogenicity of breast cancer: is it of prognostic value? Breast 2004; 13: 194–199.

Hermann

Schwartz

. Focal fibrous disease of the breast: mammographic detection of an unappreciated condition. Am J Roentgenol 1983; 140: 1245–1246.

10.

Park

Kim

. Sonographic findings of breast hamartoma: emphasis on compressibility. Yonsei Med J 2003; 44: 847–854.

11.

Mesurolle

Sygal

Lalonde

. Sonographic and mammographic appearances of breast hemangioma. Am J Roentgenol 2008; 191: W17–W22.

12.

Taboada

Stephens

Krishnamurthy

. The many faces of fat necrosis in the breast. Am J Roentgenol 2009; 192: 815–825.

13.

Linda

Zuiani

Lorenzon

. The wide spectrum of hyperechoic lesions of the breast. Clin Radiol 2011; 66: 559–565.

14.

Chao

Chen

. Sonographic features of breast hamartomas. J Ultrasound Med 2007; 26: 447–452. quiz 53.

15.

Darling

Babagbemi

Smith

. Mammographic and sonographic features of angiolipoma of the breast. Breast J 2000; 6: 166–170.

16.

Glazebrook

Morton

Reynolds

. Vascular tumors of the breast: mammographic, sonographic, and MRI appearances. Am J Roentgenol 2005; 184: 331–338.

17.

Cawson

Law

Kavanagh

. Invasive lobular carcinoma: sonographic features of cancers detected in a BreastScreen Program. Australas Radiol 2001; 45: 25–30.

18.

Watermann

Tempfer

Hefler

. Ultrasound morphology of invasive lobular breast cancer is different compared with other types of breast cancer. Ultrasound Med Biol 2005; 31: 167–174.

19.

Kobayashi

Shinozaki

Yomon

. Hyperechoic pattern in breast cancer–its bio-acoustical genesis and tissue characterization. J UOEH 1989; 11: 181–187.

20.

Lyou

Yang

Choe

. Mammographic and sonographic findings of primary breast lymphoma. Clin Imaging 2007; 31: 234–238.

21.

Glazebrook

Magut

Reynolds

. Angiosarcoma of the breast. Am J Roentgenol 2008; 190: 533–538.

22.

Bartella

Kaye

Perry

. Metastases to the breast revisited: radiological-histopathological correlation. Clin Radiol 2003; 58: 524–531.