Value of Ultrasound in Evaluation of Abnormal Axillary Lymph Node

INTRODUCTION

Detection of suspicious axillary lymph nodes (LNs) is important in breast cancer diagnosis, staging, treatment, prognosis, and follow-up. Ultrasound evaluation of axillary LNs is now a routine practice in many imaging centers, especially when the breast primary cancer is large. Although ultrasound plays an important role in detection of LN metastatic disease, many involved LNs may still be missed. It is known that overall size and shape are a relatively poor criterion for evaluation of metastasis; cortical thickness and asymmetry are more significant. The subtle sonographic signs for detecting metastatic LNs are important to know.

NORMAL SONOGRAPHIC APPEARANCE OF AXILLARY LNS

A benign LN is ovoid, with a hypoechogenic cortex, thin or even invisible at ultrasonography with a hyperechoic hilum due to connective tissue trabeculae, lymphatic tissue cords, and medullary sinusoids.^[1] Smooth thin outer echogenic cortex measuring <4 mm is a sign of a normal LN [Figures 1, 2a-c and 3].

Close

Figure 1:

An illustration of the normal anatomy of the lymph node (LN), showing the hilum, afferent and efferent vessels, capsule, follicle, cortex, medulla, artery, and vein. Metastatic cells in the lymph reach the LNs through afferent lymphatic vessels located in the convex side of the LN.

Export to PPT

Close

Figure 2a:

A 55-year-old male who presents with right retroareolar pain. LN with normal cortical thickness. 3 mm is usually considered normal.

Export to PPT

Close

Figure 2b:

A 61-year-old female who presented with left breast pain. LN with atrophic cortex and fatty infiltration of the hilum.

Export to PPT

Close

Figure 2c:

Mammographic correlate to the LN of Figure 2b.

Export to PPT

Close

Figure 3:

A 54-year-old female with history of breast cancer. Biopsy proven benign reactive (lymph node) LN. Enlarged LN with uniform cortical thickening with no transcortical vessels.

Export to PPT

BILATERAL AXILLARY ENLARGED LNs

Bilateral enlarged LNs are seen in many autoimmune diseases, for example, Rheumatoid arthritis, systemic lupus erythematosus, psoriasis, scleroderma, and dermatomyositis. They can also be seen in lymphoid hyperplasia from acute or chronic infection/inflammation, such as infectious mononucleosis and cat scratch disease, granulomatous disease such as diabetic mastopathy, sarcoidosis, tuberculosis, and silicone-induced granulomatous adenitis.^[2-4] In addition, bilateral axillary adenopathy may be seen in lymphoma and leukemia, and HIV associated conditions, for example, Kaposi sarcoma associated immune reconstitution syndrome.

It is uncommon for axillary nodal metastases from breast cancer to be bilateral, the same is true for non-breast malignancies, lung cancer, and melanoma.^[5,6]

UNILATERAL AXILLARY ENLARGED LN

Benign causes for unilateral axillary adenopathy can be mastitis or other regional infective causes such as tuberculosis or ipsilateral arm infection, for example, cellulitis and hidradenitis.^[7] Silicone or saline implants or tissue expanders may also cause axillary adenopathy if infected.

Malignant causes for unilateral adenopathy are metastatic disease from breast cancer, melanoma, or lymphoma.

[Figure 4] (Illustration of the spectrum of abnormal lymph nodes by ultrasound).

[Figures 2a-c] (Spectrum of normal lymph node appearance).

[Figure 3] (Benign appearing and favors reactive nature).

Close

Figure 4:

Illustration courtesy Dr. Don O’Conner, MD. Upper left image: Uniformly thickened cortex favors inflamed/reactive Upper middle and right images: Uneven thickening or bulbous thickening of the cortex favors metastasis Lower left: Compression of the mediastinum can be seen in metastatic or inflamed lymph node (LN). Lower middle: Absence of fatty hilum/complete nodal replacement by tumor Lower left: Classic appearance of irregular shape/extracapsular tumoral extension.

Export to PPT

Presence or absence of metastatic disease in the axilla is of crucial importance in breast cancer staging and treatment. With the increase incidence of early detection of breast cancer, the presence of axillary LNs metastasis has declined.

The choice of the method of obtaining a tissue sample of the sonographically suspicious LN whether it should be core or fine needle is sometimes a dilemma. In our institution, we believe that the core needle biopsy produces more accurate results. In a study performed by Balasubramanian et al., they also concluded that the ultrasound core needle biopsy is a superior diagnostic technique compared to the fine-needle aspiration [FNA].^[8] As to core versus FNA, we agree with the conclusion made by Ganott et al. that core biopsy can be considered if the node is clearly imaged and accessible. Fine-needle aspiration should be considered (with care to obtain sufficient material) as an alternative to core biopsy in cases of difficult accessibility or when a smaller needle is desired or other patient related factors.^[9]

We always found it important to use core needle biopsy for the patients undergoing a neoadjuvant chemotherapy as the identification of the FNA biopsied LN would be difficult after treatment, as compared to a LN with a metallic clip within.

Sentinal LN biopsy is shown to be similar to axillary dissection in metastatic detection with less morbidity in patients with early stage disease.^[10,11]

MORPHOLOGIC AND SIZE CRITERIA SUGGESTING METASTATIC NODES

Metastatic cells in the lymph reach the LNs through afferent lymphatic vessels located in the convex side of the LN. The lymph is then filtered through the cortex, paracortex and finally to the hilum. Metastatic deposits remain in the LN peripheral zone, causing enlargement of the cortex, usually small or focal in the beginning, or can be uniform^[1] [Figures 1 and 4].

Changes, such as cortical thickening, diminished or absent hilum, alterations in shape, or vascular pattern, are considered suspicious. Recent studies showed no significant relationship between size of LN and malignancy^[12,13] [Figures 5a-e and 6].

Close

Figure 5a:

A 54-year-old female with history of breast cancer. Biopsy proven metastatic reactive LN. Suspicious sign for metastasis is the presence of a normal appearing LN next to an abnormal LN with hilar loss.

Export to PPT

Close

Figure 5b:

A 56-year-old female with history of right breast cancer. Asymmetric cortical thickening favors metastasis. Biopsy proven metastatic LN.

Export to PPT

Close

Figure 5c:

A 77-year-old female with a suspicious left breast mass. Loss of normal echogenic thin cortex with fuzzy borders of the LN and some spiculations, findings favor metastasis.

Export to PPT

Close

Figure 5d:

Same patient of 3C. Complete mediastinal (hilar) obliteration favors metastatic infiltration.

Export to PPT

Close

Figure 5e:

A 60-year-old female with known left breast invasive ductal carcinoma. Indentation of the hilum in a biopsy proven metastatic LN (rat bite sign).

Export to PPT

Close

Figure 6:

A 61-year-old female who presented with right breast pain and signs of infection. Biopsy proven inflammatory/reactive lymph node (LN). The vascularity is more central than peripheral, thin echogenic capsule is preserved. Pulsed Doppler shows rounded systolic peak with a relative preserved diastolic flow and the Resistive index is not high.

Export to PPT

Stavros has also made very important points in evaluating LNs. In addition to the above parameters, he also emphasized the importance of evaluation of concentric hilar compression, convex hilar indentation “rat bite,” hilar displacement and obliteration^[14] [Figures 5a-e].

A very good parameter for evaluation of metastatic LN is the loss of the thin echogenic capsule and presence of angular margins, which suggest presence of perinodal invasion^[14] [Figure 5c].

Metastasis would affect one LN at a time, so there would be an abnormal LN adjacent to a normal appearing one, unlike inflammatory reasons which tend to affect all LNs in a regional chain^[14] [Figure 5a].

At present, many studies utilize cortical thickening and hilum absence as criteria for definition of the risk for metastasis.^[12,15-17] Absence of the hilum, rendering the LN completely hypoechoic, is the most specific alteration for metastatic disease,^[12,18] but this is only seen in advanced metastasis.

The value for cortical thickness utilized as a cutoff point for metastatic disease varies in the literature, ranging from 2.3 mm to 3 mm. Some other authors use cortex/hilum ratio.^[18,19]

A classification based on the cortical thickness has been suggested by some authors. Cho et al.,^[15] for example, have categorized the images into five grades: Grade 1, LNs with cortex ≤ 1.5 mm; Grade 2, > 1.5 and ≤ 2.5 mm; Grade 3, > 2.5 mm and ≤ 3.5 mm; Grade 4, > 3.5 mm; and intact hilum; Grade 5, > 3.5 mm and hilum absence. They have concluded that this classification is effective in the evaluation of metastasis and that cortical thickness >2.5 mm (grade three) is an indication for cytological or histological study.^[1]

Bedi et al. developed another type of classification, six types of LNs, as follows: Type 1, without visible cortex; type 2, cortex ≤ 3 mm; type 3, cortex > 3 mm; type 4, entirely lobulated cortex; type 5, with focal lobulation; and type 6, completely hypoechoic, without hilum. LNs classified as types 5 and 6 were considered suspicious; reactive LNs were frequently observed in type 3; while type 4 was considered as probably benign, since such type comprised most false negative results.^[12]

Of the known morphologic criteria for evaluation metastatic LNs, size is the least useful unless the LN is extremely and asymmetrically enlarged.^[12]

Rounding longitudinal/transverse (L/T) ratio,^[20] Cortical/ hilum (C/H) ratio.^[20] Cortical thickening >3 or 4 mm,^[15,21] Hilar effacement or loss of normal central hyperechoic hilum^[12] are all important in raising suspicion.

Performances of the CH area ratio, LT axis ratio, and blood flow pattern on power Doppler imaging for evaluation of axillary LNs.^[20] The sensitivity of the CH area ratio was superior to that of the LT axis ratio (94.1% vs. 82.3%) and to that of the blood flow pattern on the power Doppler imaging (94.1% vs. 29.4%).^[20]

Loss of echogenic thin outer cortex or angular margins can be subtle but key for suspicion.^[22,23]

DOPPLER AND POWER DOPPLER EVALUATION OF THE AXILLARY LNs

Doppler ultrasonography, of the axillary LNs follows two patterns, central pattern, and peripheral pattern. The Peripheral vascularization is more frequently found in metastatic LNs [Figures 7a and b]. Benign inflammatory/reactive LN usually shows a more central vascularity than peripheral, thin echogenic capsule is always preserved [Figure 6].^[24]

Close

Figure 7a:

A 72-year-old female with known left breast cancer. Color Doppler image of metastatic LN with transcapsular feeding vessels unlike the normal appearance of hilar vessels.

Export to PPT

Close

Figure 7b:

A 77-year-old female with a suspicious left breast mass. Metastatic LN with increased peripheral vascularity rather than the expected normal central vascularity.

Export to PPT

Malignant LNs were found to have a greater total and number of peripheral vessels compared with benign axillary LNs.^[25]

Stavros has also explored the use of color Doppler in qualitative assessment of suspicious LNs. Using color Doppler has helped in assessing angiogenesis or the formation of multiple afferent vessels that feed metastatic deposits in the cortex of the LN.^[14]

CONTRAST ULTRASOUND ENHANCEMENT

Carcinoma infiltration causes new vascular formation (neoangiogenesis) with change of the perfusion pattern with heterogeneous enhancement due to the presence of caliber changes of the neoplastic vessels and arteriovenous shunts. Malignant lymph nodes have more peripheral vessels than benign ones.^[26] They also show longer contrast enhancement duration than benign LNs. Areas of avascular necrosis (lack of contrast uptake) surrounded by hyperenhancement are another important imaging sign for malignant infiltration.

In cases of focal cortical thickening due to metastasis the cortical thickening is less well vascularized than the adjacent normal LN parenchyma compared to homogeneous enhancement in a benign LN.^[27]

In a monocenteric study performed by Agliata et al., They concluded that contrast-enhanced sonography of axillary LNs has high diagnostic accuracy and is superior to the traditional sonographic techniques. The sensitivity, specificity, and accuracy obtained by contrast-enhanced sonography for characterization of axillary LNs were 100%, 82%, and 92%, respectively.^[28]

SONOELASTOGRAPHY

Sonoelastography can improve the PPV as it offers a relative measurement of tissue stiffness. Cortex of metastatic LNs is stiffer than the cortex of healthy LNs [Figures 8a and b].^[29]

Close

Figure 8a:

A 77-year-old female with history of the left breast cancer. Biopsy proven metastatic LN with diffuse increased stiffness and lost normal echogenic hilum.

Export to PPT

Close

Figure 8b:

A 61-year-old female who presented with right breast pain and signs of infection. Biopsy proven inflammatory/reactive LN. Benign lymph node with fatty hilum and no evidence of increased stiffness.

Export to PPT

Ultrasound elastography provides valuable information for the differential diagnosis of benign and malignant LNs. The combination of elastography criteria with B-mode and Doppler-criteria may improve the PPV. Elastography can be useful in evaluation of metastatic areas in prominent LNs and to find metastatic LNs in a group of enlarged LNs with a high degree of certainty.^[30]

LYMPHOMA

In the past, lymphomatous nodes were described as hypoechoic with posterior enhancement, with a pseudocystic appearance [Figure 9]. However, with the availability of the newer high resolution transducers, lymphomatous nodes demonstrate intranodal reticulation rather than the pseudocystic appearance.^[26]

Close

Figure 9:

A 41-year-old female with history of lymphoma presents with lump in the left axilla. Biopsy proven lymphoma with increased through transmission and intranodal reticulation.

Export to PPT

Unlike metastatic nodes, lymphomatous nodes tend to have mixed vascularity and isolated peripheral vascularity is uncommon.^[26]

Lymphomatous nodes tend to be softer than metastatic nodes on elastogram.^[26]

AXILLARY LYMPHADENOPATHY AND COVID 19

During the COVID-19, pandemic a new and important cause of lymphadenopathy has emerged. A significant number of women have developed lymphadenopathy on the side of vaccine injection. The importance of knowing this is that patients with cancer should not be unnecessarily upstaged, and women without cancer should not be worked up for ipsilateral lymphadenopathy as a possible sign of occult breast cancer. Current guidelines endorsed by the Society of Breast Imaging (SBI) suggest a short-term ultrasound follow up 2–3 months after the second dose if two doses are given.^[31]

CONCLUSION

LN evaluation at the time of diagnosis of breast cancer is critical to appropriate management, either with surgery first or neoadjuvant chemotherapy whichever is indicated. Breast imaging and appropriate biopsy of axillary LNs is crucial in providing optimal care and minimizing morbidity due to unnecessary axillary surgery. For this reason, dedicated breast imagers must understand the abnormal features of involved LNs and be comfortable in providing safe sampling under ultrasound guidance. Morphologic criteria to know suggesting metastatic involvement include thickened, irregular or bulging cortex, rounding, and asymmetric hilar loss. Variability in cortical thickening in different nodes within the same axilla also favors metastasis. Transcortical vessels are abnormal and strongly suggest metastasis. Use of combined B-mode and sonoelastogram criteria of malignancy improves the specificity of ultrasound in detection of malignant LNs. This knowledge is essential for optimal care.

Unilateral axillary lymphadenopathy is currently more common than ever secondary to COVID vaccination. This should be followed with special attention to the current exceptional circumstances, and following SBI guidelines is important.