CME

Article

Sonographic Features of Gynecomastia

Vandana Dialani, MD, Janet Baum, MD, Tejas S. Mehta, MD, MPH

Objective. The purpose of this study was to identify sonographic features of gynecomastia. Methods. A retrospective analysis was performed on all male patients with breast symptoms imaged with breast sonography over a 5-year period. Breast sonograms in 158 men were jointly reviewed by 3 investigators. Sonograms were assessed for the presence or absence of a mass: (1) if mass present, (a) location of the mass, (b) vascularity, (c), axis, (d) appearance of posterior tissues, and (e) tissue echo texture; and (2) if mass absent, anteroposterior (AP) depth at the nipple (increased if >1 cm). Results. Of the 237 men with breast symptoms, 79 with only mammography were excluded. Of the 158 who had sonography with or without mammography, 5 without gynecomastia were also excluded. A total of 153 men included in the study presented with pain (n = 38), a lump (n = 95), both pain and a lump (n = 17), or nipple discharge (n = 3). Nine of 153 with gynecomastia had a biopsy. A total of 219 sonographic examinations were performed, which revealed 73 masses (33%): 20 (27%) nodular, 20 (27%) poorly defined, and 33 (45%) flame shaped. All masses were retroareolar, with 57 (78%) hypoechoic, 54 (73%) avascular, 60 (82%) parallel to the chest wall, and 47 (64%) without posterior enhancement or shadowing. Of the 146 without masses (67%), 141 (97%) had increased AP depth at the nipple. Conclusions. Gynecomastia is a clinical diagnosis, and mammography is the primary imaging modality when indicated. However, if sonography is used when mammography is declined or when mammography is inconclusive, it is important to recognize the various described patterns of gynecomastia to avoid unnecessary biopsy based on sonographic findings. Key words: breast; gynecomastia; sonography.

Abbreviations AP, anteroposterior

Received October 20, 2009, from the Department of Radiology, Beth Israel Deaconess Medical Center, Boston, Massachusetts USA (V.D., T.S.M.); and Department of Radiology, Cambridge Health Alliance, Cambridge, Massachusetts USA (J.B.). Revision requested November 16, 2009. Revised manuscript accepted for publication December 18, 2009. Address correspondence to Vandana Dialani, MD, Department of Radiology, Beth Israel Deaconess Medical Center, 330 Brookline Ave, Boston, MA 02215 USA. E-mail: vdialani@bidmc.harvard.edu
CME

Article includes CME test

ynecomastia is one of the most common diseases of the male breast.1 Palpable, painful, dense, and mobile subareolar tissue in the male breast suggests the presence of gynecomastia. One series showed palpable breast tissue in 57% of the male population older than 44 years.2 Male breast cancer is a rare entity, with incidence of less than 1%.3 There is considerable literature describing mammographic patterns of gynecomastia; however, there is very little written about the sonographic patterns. Although prepubertal gynecomastia is a clinical diagnosis,4 sonography is a documented modality of choice in evaluation of prepubertal male breast enlargement.5 However, very little is known about imaging features of gynecomastia in adults. The most commonly described sonographic pattern of gynecomastia in adults is the flame-shaped or triangular retroareolar density.6,7 A study by Wigley et al8 described 2 patterns of gynecomastia on sonography: focal, as a discrete triangular hypoechoic area in the

2010 by the American Institute of Ultrasound in Medicine J Ultrasound Med 2010; 29:539547 0278-4297/10/$3.50

Sonographic Features of Gynecomastia

retroareolar region and diffuse, with a hyperechoic increase in the amount of breast parenchyma. The limitation of the study was that it included only 4 patients. On reviewing the literature, it is unclear whether sonography is beneficial in diagnosis and whether sonography alone or in combination with mammography should be used for evaluation of gynecomastia. As a result, we decided to review the mammograms and sonograms of all symptomatic male patients referred to our department during a 5-year period.

Materials and Methods

Patients The protocol was approved by the Beth Israel Deaconess Medical Centers Institutional Review Board. Using medical records, all male patients with breast symptoms presenting for imaging over a 5-year period from July 1, 1999, to June 30, 2004, were identified, and a retrospective analysis was performed on all patients who had breast sonography. A total of 237 men presented to our radiology department with breast symptoms over a 5-year period. Of these, 158 had sonography with or without mammography, and 79 had only mammography. The 79 men with only mammography were excluded, and 5 without gynecomastia as noted on the biopsy results during data analysis (2 angiomyolipomas, 1 abscess, 1 hematoma, and 1 breast cancer) were also excluded from the study. Therefore, only 153 patients were included in the study, 66 of whom had sonography only and 87 of whom had both mammography and sonography. Study Design All sonographic examinations were performed on an HDI 5000 machine (Philips Healthcare, Bothell, WA) with a linear transducer (L10-5 or CL10-5) in a standard supine position with arm above the head. A thick layer of gel or a true standoff pad was used to better image the subareolar region. Three investigators jointly reviewed each imaging study without knowledge of the clinical history or any accompanying imaging study. The mammograms were reviewed separately from the sonograms without knowledge of the clinical or correlative sonographic
540

finding. In cases where the reviewers did not agree on the pattern, a decision was made on the basis of majority to reach a consensus. Sonograms were assessed for the presence or absence of a mass. If a mass was present, this was categorized as follows: (1) shape of the mass (nodular, poorly defined, speculated, or flame shaped); (2) location of the mass (retroareolar or other); (3) vascularity (absent, present, or not determined); (4) lesion orientation (parallel or perpendicular to the chest wall or neither); (5) appearance of posterior acoustic features (enhancement, shadowing, or neither); and (6) tissue echo texture (isoechoic, hypoechoic, or hyperechoic). If there was no mass, the anteroposterior (AP) depth of breast tissue at the nipple (from the skin surface under the nipple to the anterior surface of the pectoralis muscle) was measured and categorized as increased if the depth was greater than 1 cm. Data were analyzed for the symptomatic breast as well as the imaged contralateral breast, even if asymptomatic (contralateral breast categorized as asymptomatic). Categories Mammographic findings were categorized into 1 of 4 groups on the basis of the parenchymal pattern as described by Appelbaum et al9: dendritic, nodular, diffuse, or combination. We categorized the sonographic findings as follows: (1) nodulardiscrete round or oval hypoechoic area in the retroareolar region (Figure 1); (2) poorly definedvague hypoechoic area in the retroareolar region (Figure 2); (3) flame shapedirregular hypoechoic area with extensions into the surrounding tissue (Figure 3); and (4) nonmass lesions, subdivided as follows: (a) increased AP depth at the nipple, defined as greater than 1-cm depth of breast tissue at the nipple (which may have been isoechoic, hypoechoic, or hyperechoic; Figure 4); and (b) a normal appearance with less than 1-cm depth of breast tissue at the nipple and without any of the above criteria for a mass (Figure 5).

Results
There were 153 patients included in our study population, who presented with pain (n = 38), a lump (n = 95), pain and a lump (n = 17), or nipple
J Ultrasound Med 2010; 29:539547

Dialani et al

Figure 1. Images from a 54-year-old man who presented with a palpable tender mass in the left breast. A, A discrete oval hypoechoic mass (arrows) is shown in the retroareolar region (nipple marked as N), consistent with nodular gynecomastia. B and C, Mediolateral oblique and craniocaudal mammograms show a retroareolar mass, consistent with nodular gynecomastia.

Figure 2. Images from a 69-year-old man with known prostatic cancer who presented with right breast pain. A, A poorly defined vague hypoechoic mass (arrow) is shown in the retroareolar region (nipple marked as N), consistent with irregular gynecomastia. B and C, Mediolateral oblique and craniocaudal mammograms show retroareolar dendritic breast tissue.

J Ultrasound Med 2010; 29:539547

541

Sonographic Features of Gynecomastia

discharge (n = 3). The age range was 18 to 97 years (median age, 58 years). A total of 219 breast sonographic examinations were performed, 66 bilateral (n = 132 breasts scanned) and 87 unilateral breasts. Of the 219 breast sonographic examinations, 191 were for the symptomatic breast, and 28 were for the asymptomatic breast.

Figure 4. Images from an 84-year-old man who presented with right breast enlargement for 1 month. He had been receiving antihypertensive medication for 15 years. A, There is no discrete mass; however, the tissue depth at the nipple (N) is greater than 1 cm. This pattern of increased AP depth is another sonographic appearance of gynecomastia. B and C, Mediolateral oblique and craniocaudal mammograms show retroareolar dendritic breast tissue.

Figure 3. Images from a 42-year-old man who presented with painful nodularity in the left breast. A, An irregular hypoechoic area with extensions into the surrounding tissue (arrow) is shown in the retroareolar region (nipple marked as NIP). This is the flame-shaped appearance of gynecomastia. B and C, Mediolateral oblique and craniocaudal mammograms show retroareolar dendritic breast tissue.

542

J Ultrasound Med 2010; 29:539547

Dialani et al

These revealed 73 of 219 (33%) with masses and 146 (67%) without masses. Of the 73 with masses, 20 (27%) were nodular; 20 (27%) were poorly defined; and 33 (45%) were spiculated or flame shaped. All 73 masses (100%) were retroareolar, with 57 (78%) hypoechoic, 54 (73%) avascular, 60 (82%) parallel to the chest wall, and 47 (64%) showing no enhancement or shadowing. Of the 146 without masses, 141 (97 %) had increased AP depth at the nipple, and 5 (3%) had no increased AP depth at the nipple (normal). The largest category that showed increased AP depth at the nipple without a mass had an isoechoic tissue echo texture (76 of 141 [54%]), followed by hypoechoic (61 of 141 [43%]) and hyperechoic (4 of 141 [3%]) echo textures. The pattern of gynecomastia differed in symptomatic and asymptomatic breasts. Masses were seen in 16 symptomatic and 9 asymptomatic breasts. Increased AP depth was seen in 64 symptomatic and 15 asymptomatic breasts. Sonographic findings were normal (without gynecomastia) in 1 symptomatic and 4 asymptomatic breasts. Ninetyfive percent of patients with a nodular mass and 85% with a flame-shaped mass were symptomatic; 89% of patients with increased AP depth were also symptomatic (Table 1). In our study population there were 3 patients receiving highly active antiretroviral therapy treatment, 2 patients with a history of insulindependent diabetes mellitus, and 1 patient with a known case of Klinefelter syndrome. The 3 patients with nipple discharge in our series were receiving antiandrogens for prostate cancer, and the nipple discharge was bilateral, spontaneous, and attributed to the medication. There were no intraductal filling defects, and the discharge subsided in 3 to 5 months; hence, it was not persistent, and no further evaluation was done. Intervention (sonographically guided core needle biopsy with a 14-gauge spring-loaded biopsy device; Bard Biopsy Systems, Tempe, AZ) was performed in 9 patients. In these 9 patients with proven gynecomastia, biopsy was recommended by the radiologist: in 3 patients for eccentric densities on the mammograms and in 6 for the presence of microlobulated margins, which remained suspicious for cancer. Of the remaining 144 patients who did not have a biopsy, 2year follow-up was available for 136 patients
J Ultrasound Med 2010; 29:539547

Figure 5. Images from a 35-year-old man who presented to our department with pain in the left breast and an asymptomatic right breast. A, Normal AP thickness of less than 1 cm at the nipple (N) is shown (measured between the arrows). B and C, In the right asymptomatic breast, mediolateral oblique and craniocaudal mammograms of the right breast show a relatively fatty breast (nipple marked by a BB).

543

Sonographic Features of Gynecomastia

Table 1. Patterns of Gynecomastia in Symptomatic and Asymptomatic Patients

Sonographic Findings No Mass or Increased AP Depth

Breasts

Nodular

Irregular

Flame Shaped

Total Masses

Increased AP Depth

Symptomatic (n = 191) Asymptomatic (n = 28) Total (n = 219)

19/191 (10) 1/28 (3) 20/219 (9.1)

17/191 (9) 3/28 (11) 20/219 (9.1)

28/191 (11) 5/28 (17.7) 33/219 (15.1)

64/191 (33.5) 9/28 (32.1) 73/219 (33.3)

126/191 (67) 15/28 (53.5) 141/219 (64.4)

1/191 (0.5) 4/28 (14.3) 5/219 (2.3)

Values are number (percent).

with a clinical and imaging diagnosis of gynecomastia. Eight patients were lost to follow-up after the first visit. Mammography was performed in 87 of the 153 patients. Fifteen patients had unilateral mammograms, and 72 had bilateral mammograms (159 breasts imaged mammographically). Of the 159 breasts examined with mammography, we found 39 dendritic, 51 nodular, 24 diffuse, and 7 combination patterns. Most patients had a bilateral mammogram even if they were symptomatic on one side. Fifteen patients did not have a mammogram of the asymptomatic side. Negative (fatty) mammographic findings were seen in 38 breasts, and 35 of these 38 were asymptomatic. Only 5 of these 38 breasts with negative mammographic findings had a sonographic evaluation for comparison with the normal side, showing normal AP thickness at the nipple (Figure 5). The remaining 33 fatty breasts were not evaluated with sonography. Note that sonographic evaluations performed in 28 asymptomatic contralateral breasts showed imaging features of gynecomastia on mammography as well as sonography, indicative of subclinical gynecomastia. Of the 87 patients in whom mammography was performed, sonographically detected masses were seen in 5 with dendritic, 17 with nodular, 9 with diffuse, and 5 with combination patterns.

Discussion
Gynecomastia is a benign proliferation (hypertrophy) of ductal and glandular elements in the male breast. Physiologic gynecomastia occurs in neonates, at or before puberty, and with aging. Many cases of prepubertal gynecomastia are idiopathic (25%).4 Potential pathologic causes of gynecomastia are medications including hormones, increased serum estrogen, decreased
544

testosterone production, androgen receptor defects, chronic kidney disease, chronic liver disease, human immunodeficiency virus treatment,10 and other chronic illness. Gynecomastia as a result of spinal cord injury and refeeding after starvation has also been reported.11 Some would argue that gynecomastia is solely a clinical diagnosis, with which we agree; however, when clinical suspicion of gynecomastia is less certain, imaging is often used. Mammography is sufficient for diagnosis in most cases; however, it may be insufficient when there is asymmetric nodular gynecomastia or a cluster of subareolar ducts, which can form a convex margin simulating a mass (Figure 6). Some men refuse mammography, and sonography may be occasionally used. Knowledge of the various sonographic features and patterns of gynecomastia is especially important in these cases. Imaging findings of our patients with gynecomastia could be categorized into 1 of 4 patterns on sonography: (1) nodulardiscrete round or oval hypoechoic area in the retroareolar region (Figure 1); (2) poorly definedvague hypoechoic area in the retroareolar region (Figure 2); (3) flame shapedirregular hypoechoic area with extensions into the surrounding tissue (Figure 3); and (4) increased AP depth at the nipple, defined as greater than 1 cm depth of breast parenchyma at the nipple (which may have been isoechoic, hypoechoic, or hyperechoic; Figure 4). Carcinoma of the male breast is an unusual lesion with a frequency equaling only about 0.9% of the occurrence of female breast cancer12 and 0.2% of all malignancies in men. The peak incidence is in the fifth and sixth decades. Gynecomastia may not always be readily differentiated from carcinoma; however, a typical sonographic feature of cancer is an irregular mass with microlobulated margins (Figure 7).
J Ultrasound Med 2010; 29:539547

Dialani et al

Figure 6. Images from a 31-year-old man who presented with a hard palpable mass in the left breast. A, A discrete oval hypoechoic mass, is shown (arrows), more readily diagnostic of nodular gynecomastia. B and C, Mediolateral oblique and craniocaudal mammograms show an asymmetric mass forming a convex margin, marked with the more posterior BB, separate from the dendritic tissue under the nipple, marked with the anterior BB. Biopsy based on the mammographic interpretation of suspicious features showed gynecomastia.

Figure 7. Images from a 68-year-old man who presented with tenderness in the left breast and whose physician felt a retroareolar mass. A, A hypoechoic mass with microlobulations is shown (arrows), which is slightly taller than wide and suspicious for breast cancer (nipple marked as N). B and C, Mediolateral oblique and craniocaudal mammograms show a retroareolar mass with microlobulations (arrows), which is suspicious for cancer. In addition, strands of retroareolar dendritic tissue (arrowheads) are shown, which accompany gynecomastia. Biopsy based on the suspicious mammographic and sonographic features showed invasive breast cancer. The patient had negative breast cancer gene (BRCA) test results.

J Ultrasound Med 2010; 29:539547

545

Sonographic Features of Gynecomastia

Axillary nodal involvement is frequent in about 47% of patients.13 The lesion may be central or peripheral. The central location is frequent, and nipple involvement and ulceration of the overlying skin are common.12,14 An eccentric location of a lesion should be viewed with suspicion; rarely, some degree of eccentricity may be present in gynecomastia, although this is more pronounced in carcinoma. When correlated with lesions of similar stages in women, the overall prognosis is the same in men. Mammography should still be performed as a primary imaging modality. It better detects calcifications, especially microcalcifications, in men of an age at which cancer is a possibility. The pattern of calcifications is generally not as classic in men as in women,14 and the calcifications in male breast cancer often appear benign. We understand the drawbacks of our study. The first is that the category of increased AP depth at the nipple, defined as greater than 1 cm, was based only on our observation, without references. It may be difficult to completely differentiate gynecomastia from negative fatty breasts on the basis of sonography; however, this appearance rules out suspicion of cancer. The increased AP depth suggests the presence of gynecomastia in patients who have a small amount of retroareolar breast tissue seen on mammography (Figure 4), which cannot be delineated as a retroareolar hypoechoic area on sonography because of nipple shadowing, and these only show the increased thickness at the nipple. If patients only have fatty tissue, which flattens out, the hypothesis is that it presents with less thickness under the nipple (Figure 5). We acknowledge that the arbitrary value of greater than 1 cm for defining increased AP depth needs validation and are working on a study to gather more data. A second limitation was that not all patients in our study were imaged with both mammography and sonography (66 of 153 patients had only sonographic evaluations). The protocol followed was that if a mammogram was not requested or the patient declined or was resistant to mammography, breast sonography was performed as the initial imaging test. If the sonography showed no suspicious findings or findings suggestive of gynecomastia, mammography was not performed. None of these patients had new or wors546

ening breast symptoms over a 2-year clinical follow-up. Many of these patients were evaluated by our experienced radiologists (>10 years in breast imaging) with expertise in sonography, who were thus confident in identifying and distinguishing sonographic features of gynecomastia with more certainty than others with less experience and expertise. After reviewing data from this study, we have standardized our protocol to obtaining a mammogram as the initial imaging examination in all male patients older than 30 years with breast symptoms. Once the mammogram is reviewed by the radiologist, sonography is performed if necessary for diagnostic clarification. Sonography is used as the initial examination for patients younger than 30 years. This protocol is what we recommend. In conclusion, gynecomastia is a clinical diagnosis, and mammography is the primary modality when imaging is indicated. However, if sonography is used as a primary imaging tool when mammography is declined or in combination with mammography, and when mammography is inconclusive, it is important to recognize the various described patterns of gynecomastia to avoid unnecessary biopsy based on the sonographic findings. A sonographic mass can be seen in up to 33% of gynecomastia cases. The presence of a taller-than-wide lesion with microlobulations or an eccentric lesion away from the retroareolar region should raise the suspicion for breast cancer.

References
1. Gnhan-Bilgen I, Bozkaya H, Ustn EE, Memis A. Male breast disease: clinical, mammographic, and ultrasonographic features. Eur J Radiol 2002; 43:246255. Nuttall FQ. Gynecomastia as a physical finding in normal men. J Clin Endocrinol Metab 1979; 48:338340. Kuroi K, Toi M. Male breast cancer [in Japanese]. Gan To Kagaku Ryoho 2003; 30:599605. Einav-Bachar R, Phillip M, Aurbach-Klipper Y, Lazar L. Prepubertal gynaecomastia: aetiology, course and outcome. Clin Endocrinol (Oxf) 2004; 61:5560. Welch ST, Babcock DS, Ballard ET. Sonography of pediatric male breast masses: gynecomastia and beyond. Pediatr Radiol 2004; 34:952957. Dershaw DD. Male mammography. AJR Am J Roentgenol 1986; 146:127131.

2. 3. 4.

5.

6.

J Ultrasound Med 2010; 29:539547

Dialani et al

7.

Stewart RA, Howlett DC, Hearn FJ. Pictorial review: the imaging features of male breast disease. Clin Radiol 1997; 52:739744. Wigley KD, Thomas JL, Bernardino ME, Rosenbaum JL. Sonography of gynecomastia. AJR Am J Roentgenol 1981; 136:927930. Appelbaum AH, Evans GF, Levy KR, Amirkhan RH, Schumpert TD. Mammographic appearances of male breast disease. Radiographics 1999; 19:559568. Peyriere H, Mauboussin JM, Rouanet I, et al. Report of gynecomastia in five male patients during antiretroviral therapy for HIV infection. AIDS 1999; 13:21672169. Heruti RJ, Dankner R, Berezin M, Zeilig G, Ohry A. Gynecomastia following spinal cord disorder. Arch Phys Med Rehabil 1997; 78:534537. Crichlow RW, Galt SW. Male breast cancer. Surg Clin North Am 1990; 70:11651177. Mathew J, Perkins GH, Stephens T, Middleton LP, Yang WT. Primary breast cancer in men: clinical, imaging, and pathologic findings in 57 patients. AJR Am J Roentgenol 2008; 191:16311639. Dershaw DD, Borgen PI, Deutch BM, Liberman L. Mammographic findings in men with breast cancer. AJR Am J Roentgenol 1993; 160:267270.

8.

9.

10.

11.

12. 13.

14.

J Ultrasound Med 2010; 29:539547

547