Abstract
The presence of mammographically evident hyperdense foci within axillary lymph nodes elicits concern for calcium deposits, which in turn have a wide differential diagnosis including both benign and malignant entities. Tissue sampling, most commonly by way of image-guided core needle biopsy, is needed in many cases when a definite etiology cannot be clinically established. In this case series we present history, imaging findings, and pathology results (or long term follow-up stability as biopsy surrogate) of several women with body tattoos who at mammography were noted to have a characteristic pattern of “bubbly” pseudo-calcifications within axillary lymph nodes, and absence of other mammographic, sonographic and clinical abnormalities.
Case 1
History
A 37-year-old woman (Fig. 1) who presented for a baseline mammogram due to bilateral breast palpable abnormalities with associated breast tenderness. She denied any history of systemic disease such as sarcoidosis, histoplasmosis, treated lymphoma or gold injections. On clinical inspection, she had extensive tattooing of the back and bilateral arms.
37-year-old woman who presented with bilateral palpable and tender breast abnormalities due to simple and complicated cysts on mammographic and sonographic evaluation (ultrasound images of cysts not shown). The patient had a large polychromatic tattoo covering her back and bilateral arms (not shown). a. Bilateral synthetic 2D MLO views demonstrate multiple, bilateral normal sized axillary lymph nodes with extensive, hyperdense cortical “bubbly” calcifications (arrows indicate dominant nodes). b. Gray-scale US images demonstrate right and left axillary lymph nodes (arrows) with normal size, shape and cortical thickness without hyperechoic foci to suggest calcifications. c. Mammographic BB localization of a representative LN using an alpha-numeric grid. d. Gray-scale US confirmed the LN of concern. e. Gray-scale image of ultrasound guided core biopsy using a 14g need and BARD Marquee disposable automated gun with open aperture technique. f. High resolution specimen digital x-ray demonstrated multiple cores with hyperdensities in select cores (arrows). g. Photograph of the cores demonstrate dark violet pigment matching the areas of radiographic hyperdensities (arrows). h. Hematoxylin and eosin slides from core needle biopsy of lymph node tissue, associated with dark pigmentary alterations (arrows) at 4x (low power). The pigment infiltration pattern is curvilinear and “bubbly”.
Mammographic and sonographic findings revealed that the areas of clinical concern were due to benign simple and complicated cysts (not shown). Mammographic images demonstrated multiple small, bilateral and symmetric axillary LNs with extensive cortical “calcifications” with curvilinear or bubbly appearance (Fig. 1a). Ultrasound (US) of the bilateral axillae (Fig. 1b) failed to identify any abnormal LNs, including any echogenic foci within the LN cortices to explain the mammographic findings thought to be due to calcifications. The decision was made to perform an ultrasound guided core biopsy of a representative LN. To accurately identify one of the “calcified” nodes, the patient was placed in the alpha-numeric grid for a mammographic BB localization using a lateral to medial approach (Fig. 1c). She was then taken to the US suite, and the corresponding LN was identified (Fig. 1d). Ultrasound-guided core biopsy was performed with a 14-gauge automated gun (BARD, Marquee 14 gauge) (Fig. 1e). Radiograph of the specimen cores showed hyperdense foci (Fig. 1f).
Pathology findings
Macroscopically, several cores demonstrated dark violet and black pigment (Fig. 1g), which correlated with the extent and distribution of the hyperdense foci on high resolution x-ray of the specimen cores. Histopathologic examination of the biopsy cores yielded LN tissue with extensive dark brown to black, coarse granular pigment in a paracortical and sinusoidal distribution (Fig. 1h), a pattern that correlates with the “bubbly” appearance of the hyperdensities on mammographic images.
Case 2
History
A 56-year-old woman with a history of a left breast central lumpectomy due to invasive ductal carcinoma, estrogen and progesterone receptor positive, and human epidermal growth factor receptor 2 negative, 10 years prior, with 4 negative sentinel lymph nodes. A left nipple tattoo was performed as part of her left breast reconstruction, which she reported had recently faded significantly.
Imaging findings
On mammography, new right (contralateral to the prior breast cancer) axillary LN curvilinear “bubbly” cortical calcifications were identified in the lower pole of an otherwise mammographically stable LN (Fig. 2a). There were no other suspicious mammographic abnormalities in either breast. US of the right axilla showed a morphologically normal LN without cortical thickening or calcifications (Fig. 2b). Because of the history of contralateral cancer, a biopsy of the lower pole of the LN was performed using a 12-gauge ATEC Vacuum assisted device. Hyperdensities were identified on biopsy specimen digital radiograph (Fig. 2c).
A 56-year-old woman with a history of a left central lumpectomy due to IDC: ER/PR (+), HER2 (−) 10 years prior, with 4 negative sentinel lymph nodes. A left nipple tattoo was performed as part of her left breast reconstruction, which she reported had recently faded significantly. a. Cropped right synthetic 2D MLO view showing new calcifications within the lower pole of a fatty replaced axillary LN (arrow). b. Gray-scale images of the corresponding morphologically normal LN without calcifications in the lower pole. c. Specimen digital x-ray with hyperdensities within several cores (arrows). d. Photograph of specimen container with hyperpigmented cores following the distribution of the x-ray finding (arrows). e. Hematoxylin and eosin slide from core needle biopsy of lymph node tissue, associated with dark pigmentary alterations (arrows) at 20x.
Macroscopically, several cores demonstrated dark purple pigment (Fig. 2d), corresponding to hyperdensities on biopsy specimen radiograph. Histopathology showed presence of tattoo pigment within the cores, without calcifications or malignancy (Fig. 2e).
Case 3
History
42-year-old asymptomatic woman with large upper torso and right extremity tattoos, presented for baseline screening mammogram.
Imaging findings
Mammographic evaluation was negative, except for right axillary LN “bubbly” calcifications (Fig. 3a). US of the right axilla demonstrated fine linear echogenic foci (Fig. 3b), which were targeted on subsequent ultrasound guided core needle biopsy. Specimen digital radiograph of the biopsy core specimens showed hyperdense foci (Fig. 3c).
42-year-old asymptomatic woman who presented for baseline routine mammogram. She had a large upper torso and extremity tattoo. a. Cropped synthetic 2D right MLO view demonstrates a left axillary node with cortical “bubbly” calcifications (arrow). b. Gray-scale US showed fine echogenic linear foci without posterior acoustic shadowing (arrow). c. Specimen digital x-ray of cores obtained by US guided biopsy yielding specimens which contained dark violet pigment (not photographed) on clinical inspection, reported as “benign reactive lymph node”. Calcifications were not identified despite performing paraffin blocks. d. Hematoxylin and eosin slide from core needle biopsy of lymph node tissue, associated with dark pigmentary alterations (arrows) at 20×, which pathology confirmed after informed of the presence of a tattoo.
Macroscopically, the biopsy tissue cores demonstrated dark violet pigment. Histopathologic examination yielded benign, reactive lymph node tissue without calcifications, which were also not seen on x-ray of the paraffin block (not shown). The presence of pigment was reported on pathology (Fig. 3d).
Case 4
History
A 43-year-old asymptomatic woman presented for routine screening mammogram. She reported getting a new tattoo on her right arm a few months prior to this mammogram.
Imaging findings
On mammography there were new “bubbly” calcifications (Fig. 4a) within an otherwise stable 1 cm round and dense right axilla lymph node compared to a screening mammogram from 8 years prior (Fig. 4b). Ultrasound of the right axilla (Fig. 4c) confirmed normal lymph node morphology, without cortical thickening and absence of calcifications. No LN biopsy was performed and she remained mammographically stable (not shown) for another 3 years after the initial presentation of the bubbly calcifications.
43-year-old asymptomatic woman who presented for screening mammogram. She acquired a right arm tattoo a few months prior. a. Cropped Right synthetic 2D MLO projection shows new “bubbly calcifications” within a round, dense, right axillary lymph node (arrows). b. The LN (arrow) was stable in size and shape compared to a mammogram obtained 8 years prior. c. Gray-scale US image with normal LN (arrow) without cortical echogenic foci.
History
Six asymptomatic patients without a personal history of breast, ovarian or thyroid malignancy or any other known benign cause of axillary LN calcifications, five of whom had significant torso or upper extremity tattoos presented for routine mammography. The final patient (Fig. 5f) had recently undergone permanent make up/tattoos of the eyebrows or “microblading”. These patients were not biopsied, but had imaging follow up ranging from 2 to 6 years (average of 2.5 years) after the initial presentation of “calcifications” without a new cancer diagnosis.
Cropped and digitally magnified MLO views of 6 different women (ages ranging from 39 to 54 years without a personal history of malignancy) with characteristic “bubbly cortical calcifications” in axillary LNs (arrows), who presented without mammographic or sonographic abnormalities in the breasts. No abnormal LN US features or cortical echogenic foci were identified. All presented with upper torso or extremity tattoos, except for 5f who acquired eyebrow tattoos (microblading) less than a year prior to this mammogram (bilateral cortical calcifications were seen, but only the right were shown).
55-year-old woman who presented with a palpable right upper outer quadrant mass with associated skin retraction and ulceration. Biopsy of the right breast mass yielded IDC, ER/PR+, HER 2 (−) breast cancer. The axillary lymph node biopsy confirmed metastatic adenopathy. a. Right 2D MLO projection demonstrates marked skin retraction and thickening at the site of palpable concern and ulceration, denoted by a triangular marker. There is global asymmetry in this area with grouped coarse heterogenous calcifications. The arrow demonstrates a low-lying completely calcified axillary lymph node with coarse heterogeneous calcifications similar to those in the breast lesion. b. Cropped magnified right MLO view showing the same as figure a. c. Gray-scale US shows a hypoechoic, ill-defined and completely calcified round mass in the right axillary tail consistent with the calcified LN. d. High resolution, 2D magnification view of specimen digital x-ray of the lymph node cores demonstrates multiple cores containing hyperdense calcifications (representative core annotated with arrow).
48-year-old woman who presented with left breast focal pain. Bilateral digital mammogram demonstrated four equal density, non-calcified irregular masses with associated architectural distortion, (the largest and most superficial of which was the cause of pain) and 2 abnormal axillary LNs with associated cortical calcifications. Subsequent biopsy of the 2 dominant breast lesions and dominant axillary LN yielded IDC: ER/PR+, HER 2 (−) and metastatic adenopathy. a. Cropped and digitally magnified left synthetic 2D MLO view demonstrates four irregular, equal density masses which ranged in size from 1.6 to 0.9 cm with associated architectural distortion in segmental distribution at approximately 2 o’clock (encircled in ellipse). Arrows demonstrate the dominant, dense, replaced left axillary lymph nodes with faint, amorphous microcalcifications. b. Gray-scale US image shows a representative LN with hypoechoic cortex, compressed hilum and multiple echogenic foci involving the cortex and hilum (arrow). c. Gray-scale US image shows a round hypoechoic LN with compressed hilum and cortical calcifications (arrow).
59-year-old woman with a history of stage IV high grade serous ovarian adenocarcinoma, otherwise asymptomatic. She presented for a screening mammogram as a prerequisite for a clinical trial. Screening mammogram showed no abnormalities in either breast. a. Cropped and digitally magnified left synthetic 2D MLO view shows dense, rounded axillary lymph nodes, one of which contains faint, amorphous, peripheral calcifications (arrow). b. Gray-scale US demonstrates a round, hypoechoic, completely replaced axillary lymph node with easily identified cortical microcalcifications (arrows). c. Magnified specimen digital x-ray demonstrates multiple cores of the left axillary LN biopsy. Arrow denotes the dominant core with calcifications.
Mammographic evaluation for each of the six patients was negative except for identical axillary LN curvilinear or “bubbly” calcifications within otherwise morphologically normal LNs (Figs 5a through 5f). They each had normal axillary US, without suspicious morphology or evidence of cortical echogenic foci (not shown).
Differential diagnosis of axillary lymph node calcifications
Differential diagnosis of axillary lymph node calcifications
Algorithm for management of new axillary lymph node “calcifications” on mammography.
The presence of axillary lymph node (LN) hyperdense foci on mammography is a cause for concern, as the differential diagnosis includes metastatic breast carcinoma [1], as well as metastatic ovarian and papillary thyroid carcinoma [2,3] with psammomatous calcifications, often requiring a confirmatory biopsy.
With an increasing prevalence of body art in the form of tattoos which have progressed in size, extent, and array of pigments, it is important for radiologists involved in breast imaging to be aware of the associated mammographic and sonographic manifestations of this phenomenon, particularly its effect on the axillary lymph nodes (LNs) which are often visualized on Medio Lateral Oblique (MLO) projections on routine mammograms. The typical mammographic presentation consists of unilateral axillary LNs which are normal in size, shape and density, with associated hyperdense cortical foci initially interpreted as calcifications with frequently normal appearance on ultrasound (US). Occasionally, LN hyperdense foci may be bilateral, depending upon the distribution and extent of the body art, most commonly involving the upper extremities or trunk.
The overall prevalence of tattooed people in five major countries is 18.5%. Tattoo prevalence ranges from 11.7% in Russia to 31.5% in the USA, and Americans are more likely to have greater than one tattoo compared with other countries. Overall, women are more often tattooed than men (19.4% vs. 17.6%) [4]. A national survey performed in Germany demonstrated that 83.7% of individuals with tattoos were between the ages of 15–49, with 11.9% between the ages of 50 and 64 [5]. Nipple tattoos as part of the reconstruction process of patients undergoing simple mastectomy, skin sparing mastectomy and central lumpectomy have also become more frequent [6], and permanent make up in the form of eyebrow tattoo or microblading has become increasingly popular [7,8].
The dyes used for tattoos include metals such as aluminum, titanium, iron, mercury sulfide, cobalt, zinc, chromium, manganese, barium sulfate and iron as well as non-metallic substances such as carbon and curcuma [9,10]. These elicit an inflammatory response with pigment absorption by macrophages which eventually drain to and deposit within the regional lymph nodes [11]. For those tattoos involving the upper torso and extremities, these may include the axillary lymph nodes.
Several single case reports [12–19] have addressed the radiographic findings of hyperdense foci simulating calcifications within axillary nodes on mammography, attributed to the deposition of tattoo pigment. To our knowledge, this is the largest case series presented. In addition to describing 8 patients with characteristic body tattoos, our case series includes 2 unusual cases: the first involving tattoo pigment from a contralateral nipple tattoo as part of the reconstruction process after a central lumpectomy, performed 10 years prior to presentation of the LN hyperdensities. In this case the contralateral axillary deposition of the pigment may be explained by previously reported aberrant lymphatic drainage to the contralateral (right) axilla due to the left sentinel lymph node procedure [20,21]. The second case involves recent eyebrow tattoos or “microblading”, not previously reported; these findings are unusual given the literature description of the “semipermanent” nature of the pigments used for this purpose [7,8]. A wide range of the time interval between acquiring the tattoo and mammographic presentation of the LN findings ranging from a few months to 10 years was found in this series.
The distinct mammographic pattern of curvilinear or “bubbly” pattern and cortical distribution in the LNs with tattoo pigment has not yet been described, although a single report of a “curvilinear” cluster of dermal microcalcifications corresponding to the shape and location of a tattoo on the breast confirmed on tangential imaging was reported [22]. In this limited case series, all patients presented with identical characteristic curvilinear or “bubbly” cortical calcifications. In the LNs, this pattern mirrors that of infiltration of the tattoo pigment within the LN cortex, described on histopathology as paracortical and sinusoidal deposition of dark, coarse, granular pigment, forming a “bubbly” or curvilinear distribution which can be seen on histology (Figs 1h, 2e and 3d).
The frequent lack of corresponding abnormal morphologic features of the lymph nodes affected by tattoo pigment on mammography and US has not been previously emphasized. LNs should present normal reniform shape with normal cortical thickness and echogenicity on US with hilar preservation. More importantly, the lack of “calcifications” on US has only rarely been described in the literature [16,19]. This may be due to the tiny size of the dye particles or their specific components which may not produce echogenic effects or posterior acoustic shadowing on US, as typically seen with calcium deposits. Only 1 of the 10 patients in our series (10%) demonstrated sonographic evidence of corresponding echogenic foci, although within an otherwise morphologically normal LN. In the setting of malignant calcified LNs, other abnormal lymph node characteristics are identified mammographically and on ultrasound such as increase in size and density, loss of reniform shape, irregular margins, changes in the cortical thickness, or hypoechoic cortex on ultrasound. The malignant LN calcifications may be pleomorphic or coarse heterogeneous, most frequently seen in breast cancer (Fig. 6), but may also appear faint and amorphous (Figs 7 and 8). In addition, ultrasound can identify cortical echogenic foci with or without posterior acoustic shadowing which confirms the presence of calcium that can be targeted on ultrasound guided biopsy.
For tattoo pigment to be considered the cause of axillary LN hyperdense foci, the patient must have a characteristic tattoo as described above (including nipple tattoos and eyebrow microblading), and have no signs, symptoms or imaging findings concerning for breast malignancy such as palpable breast abnormalities, skin thickening, nipple retraction or discharge. There should be no history of a malignancy known to cause calcified axillary metastasis, such as breast [1], ovarian [2] or thyroid cancers [3]. In addition, non-malignant differential diagnosis including granulomatous processes such as tuberculosis (and BCG vaccination), sarcoidosis and histoplasmosis as well as gold injections for treatment of arthritis, fat necrosis [23–26] and silicone deposits should be excluded. Calcified adeonpathy secondary to treated lymphoma should be considered as well [27,28]. In these cases, the patient’s medical history can usually elicit an appropriate diagnosis. See Table 1 for a detailed description of differential diagnosis of axillary LN hyperdense foci on mammography and their imaging characteristics.
The presence of clinically evident dark violet or black pigment within cores obtained during an axillary LN biopsy (with or without the presence of calcifications on mammography) can also be attributed to tattoo pigment. In this scenario, the differential diagnosis includes metastatic melanoma [29,30] and dermatopathic lymphadenopathy [31] in which benign reactive LN hyperplasia consisting of macrophages which contain melanin pigment as a response to increased antigenic stimulation in the skin occurs. This may be caused by conditions such as pemphigus, psoriasis, eczema, mycosis fungoides or Sezary syndrome and autoimmune processes such as systemic lupus erythematous (SLE).
Although this is a limited case series, there is a growing body of evidence in the literature of the frequency of these axillary LN findings on routine breast imaging in otherwise asymptomatic patients who have tattoos and no risk factors for malignant etiologies. We propose the following algorithm for management of patients with the characteristic “bubbly” LN calcifications that takes into account a patient’s history, mammographic and sonographic findings, and the lack of abnormal LN features on US, in which case imaging surveillance may be considered in lieu of biopsy (Fig. 9).
Conclusion
Tattoo pigment may be considered the cause of mammographic axillary lymph node cortical pseudo-calcifications or hyperdense foci with a “bubbly” appearance in patients who have an upper body tattoo, a normal axillary ultrasound (including the absence of cortical echogenic foci) and who have no history or imaging features to suggest breast, ovarian, or thyroid malignancy. In these patients, consideration may be given to imaging surveillance in lieu of lymph node biopsy, resulting in reduced anxiety and cost.
Conflicts of interests
None of the authors have any conflicts of interest to disclose.
Abbreviations
Medio Lateral Oblique (MLO)
Lymph nodes
Lymph node
Ultrasound
Gauge
Invasive Ductal Carcinoma
Estrogen
Progesterone
Human Epidermal Growth Factor Receptor
Systemic Lupus Erythematous
Figure