Benign vs. Malignant Neck Lymph Node Ultrasound Features

On ultrasound, benign neck lymph nodes tend to be oval, show a bright central stripe called an echogenic hilum, and carry blood flow through that central core. Malignant nodes, by contrast, tend to be rounder, lose their hilum, develop internal necrosis or calcifications, and draw blood supply from the edges inward. No single feature is a perfect giveaway, though, and the real diagnostic power comes from reading several features together, sometimes with the help of newer tools like elastography and contrast-enhanced imaging.

What a Normal or Reactive Node Looks Like

A healthy cervical lymph node is oval or bean-shaped, hypoechoic (darker than surrounding tissue), and has a visible echogenic hilum running through its center. That hilum is a fatty, fibrous zone where blood vessels enter the node, and its presence is one of the most reassuring signs a sonographer can see. Normal nodes also tend to have either no detectable blood flow on Doppler or a flow pattern that branches neatly from the hilum outward.¹

Reactive nodes, the kind that swell when you have a sore throat or a dental infection, look almost identical to normal nodes on ultrasound. They are still oval, still hypoechoic, and still show a preserved hilum. Their blood flow is predominantly hilar, just sometimes more prominent because of the inflammation driving more blood into the node. This overlap is important: a swollen node that keeps its normal architecture is unlikely to be cancer, even if it is bigger than expected.

Shape and hilum status are broadly consistent across ages and body types. A study of healthy Turkish adults found that normal cervical lymph nodes were oval with an echogenic hilum regardless of age, sex, or body mass index.²

Shape and Short-Axis Diameter

Round nodes are more suspicious than oval ones. A useful mental shortcut is the short-to-long axis ratio: a normal node is elongated like a kidney bean, so its short axis is much smaller than its long axis. When a node becomes rounder, that ratio climbs toward 1, and the node starts to look more like a ball than a bean. In a study comparing benign and malignant cervical nodes, the short-axis diameter was significantly larger in malignant nodes, whereas the long-axis difference was not significant.³ In other words, cancer makes nodes fatter, not necessarily longer.

Among patients with papillary thyroid carcinoma, a round shape was seen in about 80% of metastatic nodes but only about 30% of benign ones.⁴ That said, submandibular and parotid nodes can be naturally rounder even when healthy, so location matters when judging shape.

Loss of the Hilum and Echogenicity Changes

The echogenic hilum is the single most talked-about feature in benign-versus-malignant assessment. When tumor cells infiltrate a lymph node, they typically replace the normal fatty hilum from the periphery inward. On ultrasound, this looks like the bright central stripe fading or disappearing entirely. In the same papillary thyroid study, absence of the hilum was noted in about 88% of metastatic nodes but only 10% of benign ones.⁴

Echogenicity itself can shift. Normal lymph nodes are hypoechoic relative to the surrounding muscle. Metastatic nodes from thyroid cancer, however, often become hyperechoic (brighter than muscle), a feature seen in roughly 86% of metastatic nodes in that population compared with under 5% of benign ones. This is thought to reflect the thyroid tissue character of the metastatic deposit. Other primary cancers, such as squamous cell carcinoma, may not produce the same hyperechogenicity, so interpreting brightness always depends on the clinical context.

Intranodal Necrosis, Cystic Change, and Calcifications

When tumor deposits outgrow their blood supply inside a node, patches of necrosis develop. On ultrasound, necrosis shows up as dark (anechoic or hypoechoic) areas within the node, sometimes with a cystic, fluid-filled appearance. These areas lack blood flow on Doppler, which helps distinguish them from a normal hilum. Intranodal necrosis and calcification are among the grey-scale features most closely linked to malignancy.⁵

Calcifications appear as bright pinpoint or coarse echogenic foci, sometimes with acoustic shadowing behind them. They are especially common in metastatic nodes from papillary thyroid cancer, where roughly half of metastatic nodes showed calcifications compared to none of the benign nodes in one study.⁴ Microcalcifications in a cervical node should prompt close attention to the thyroid gland itself.

In papillary thyroid cancer with extranodal extension (tumor breaking through the capsule of the node into surrounding tissue), cystic areas and node matting (multiple nodes stuck together) were found to be independent risk factors for that extension.⁶ Matting, calcification, and a short-to-long axis ratio approaching 1 were also more common in that group. Seeing a cluster of nodes fused together with cystic areas inside is a worrisome combination that raises concern not just for cancer but for locally advanced disease.

How Blood Flow Patterns Help

Grey-scale features alone are not always decisive, which is where Doppler imaging adds value. The basic principle: benign and reactive nodes typically show blood flowing in from the hilum and branching outward in a tidy, symmetric pattern. Malignant nodes tend to recruit abnormal vessels from their periphery, producing flow at the edges, scattered flow throughout the node, or a mix of peripheral and central signals.

Color Doppler studies have consistently found that central or hilar flow is characteristic of benign nodes, while peripheral or mixed flow patterns are characteristic of malignant ones.⁷ An older but often-cited study observed normal vascular patterns in 94% of reactive nodes, while at least one abnormal vascular pattern appeared in 98% of nodes with malignant disease.⁸ Those abnormal patterns included eccentric or absent hilar flow, deformed central vascularity, aberrant multifocal vessels, and peripheral vascularity.

Beyond the location of flow, the character of the waveform matters. Spectral Doppler measures like the resistive index and pulsatility index tend to be higher in malignant nodes. In one study, metastatic nodes showed a resistive index above 1.0 and pulsatility index above 1.5 in the majority of cases, while all benign nodes had a resistive index below 0.8 and a pulsatility index below 1.5.⁹ Another study confirmed that the highest pulsatility and resistive index values in metastatic nodes exceeded those in benign nodes, though the lowest values overlapped.¹⁰ That overlap is a reminder that spectral Doppler adds information but does not replace the overall picture.

Elastography Adds a Stiffness Dimension

A relatively newer ultrasound tool, elastography measures how stiff tissue is. The underlying logic is straightforward: malignant nodes, packed with dense tumor cells and often fibrotic, tend to be stiffer than benign ones, which remain soft and pliable.¹¹

In shear wave elastography, the scanner sends a small push-pulse into the tissue and tracks how fast the resulting wave travels; stiffer tissue transmits the wave faster. A study comparing metastatic and benign cervical nodes found that the maximum stiffness value was significantly higher in metastatic nodes (averaging about 41 kPa) than benign ones (about 14 kPa). At a cutoff of roughly 19 kPa, the technique achieved about 91% sensitivity and 97% specificity, slightly outperforming conventional grey-scale ultrasound in that cohort.¹²

A more recent prospective study found that benign lymph nodes had a maximum shear wave velocity of 35 kPa, while lymphoma nodes started at 40 kPa and metastatic nodes at 50 kPa. Each additional 1 kPa of stiffness was associated with a 3% increase in the risk for metastatic disease.¹³ Elastography is particularly useful when grey-scale findings are equivocal, such as when a node has lost its hilum but is not clearly round or necrotic.

Contrast-Enhanced Ultrasound

Contrast-enhanced ultrasound (CEUS) uses microbubble contrast agents injected intravenously to map blood flow at the capillary level. Benign nodes tend to enhance homogeneously and show centrifugal perfusion, meaning the microbubbles spread outward from the hilum in an orderly fashion. Malignant nodes more often show inhomogeneous enhancement, perfusion defects (areas the microbubbles do not reach), and centripetal or mixed perfusion patterns.¹⁴

A meta-analysis pooling results from multiple studies confirmed that inhomogeneous enhancement, early impregnation, perfusion defects, and peripheral or heterogeneous hyperperfusion patterns were reliable features of malignant nodes.¹⁵ CEUS is not yet a routine part of every neck ultrasound, but it is increasingly used in cancer centers where a close call on Doppler needs more information before biopsy.

Why Infection and Tuberculosis Can Fool You

The biggest source of confusion on cervical ultrasound is not the difference between an obviously normal node and an obviously metastatic one. It is the murky middle ground where infectious or granulomatous disease produces features that overlap with cancer. Tuberculous lymphadenitis is the classic example. TB nodes can lose their hilum, show intranodal necrosis and cystic areas, mat together, and display peripheral vascularity on Doppler, hitting nearly every checkbox for malignancy on the list above.¹⁶

Key ultrasound features that lean toward TB include strong internal echoes, thin echogenic layers within the node (thought to reflect caseous necrosis), and soft-tissue edema around the node with displaced hilar vascularity.¹⁷ Still, none of these reliably exclude cancer. In areas where TB is common, tissue sampling is often necessary to settle the question.

Vascular patterns from the earlier color Doppler study are also worth recalling here: all tuberculous nodes in that cohort showed abnormal vascularity patterns indistinguishable from malignant disease on Doppler alone.⁸ This is a genuine clinical trap, and it is why experienced sonographers never rely on a single feature.

Lymphoma Looks Different from Metastatic Carcinoma

Lymphoma infiltrates nodes differently than a solid-organ cancer that has metastasized. Metastatic carcinoma tends to invade from the subcapsular sinus inward, producing focal cortical thickening, eccentric hilum displacement, and eventually necrosis. Lymphoma, by contrast, tends to replace the entire architecture more uniformly. On ultrasound, lymphomatous nodes are often markedly hypoechoic, round, and homogeneous, without the focal necrosis or calcifications that typify many solid-tumor metastases.

A preliminary study of children with suspicious lymphadenopathy described a pattern researchers nicknamed the “truffle sign”: enlarged nodes with loss of internal structure, mostly hypoechoic, with fine echogenic serpentine lines surrounding hypoechoic pseudonodular areas, resembling the inside of a black truffle. In nine of twelve cases showing this pattern, biopsy confirmed lymphoma. While the sample was small, the observation highlights that lymphoma has its own ultrasound signature distinct from metastatic carcinoma.

Structures That Mimic Lymph Nodes

Not every suspicious-looking structure in the neck is actually a lymph node. Partially visible or compressed veins, accessory salivary tissue, neuromas, carotid body tumors, lateral neck cysts, and even postoperative thyroid remnants can all be mistaken for abnormal nodes. Conversely, metastatic nodes can sometimes look like cysts because extensive necrosis makes them appear almost entirely fluid-filled.¹⁸

An important overall summary of the features sonographers watch for includes round shape, absent hilum, intranodal necrosis, reticulation, calcification, matting, surrounding soft-tissue edema, and peripheral vascularity. No single feature is diagnostic by itself, but combinations are powerful.¹⁹

Scoring Systems and Risk Stratification

Because no single ultrasound feature is perfectly sensitive or specific, researchers have built scoring systems that combine multiple features into a single risk score. For papillary thyroid cancer, one model that integrated suspicious ultrasound features achieved strong predictive performance with a concordance index of 0.84.²⁰ For head and neck squamous cell carcinoma, a separate scoring system reached sensitivity of about 74% and specificity of about 95% at a cutoff of 3 points.²¹

These systems are designed to standardize what can otherwise be a subjective call. A node might look borderline suspicious to one radiologist and benign to another. Scoring removes some of that variability by assigning points for each worrisome feature (round shape, lost hilum, necrosis, abnormal vascularity, and so on) and setting a threshold for recommending biopsy.

Where AI Fits In

Machine learning tools trained on ultrasound images are increasingly being tested for lymph node classification. A systematic review and meta-analysis of ultrasound-based AI models found pooled sensitivity of about 84% and specificity of about 85% for distinguishing benign from malignant nodes.²²

For the specific task of predicting cervical lymph node metastasis in papillary thyroid cancer, AI models showed higher sensitivity (about 80%) than human ultrasound physicians (about 51%), though specificity was similar for both.²³ The gap was widest in sensitivity, meaning AI was better at catching metastatic nodes that humans missed, while both performed comparably at correctly identifying benign ones. These tools are not yet standard clinical practice, but they are moving in that direction as decision-support aids rather than standalone diagnostics.

Pediatric Cervical Nodes on Ultrasound

Children normally have palpable cervical lymph nodes, and their nodes tend to be larger and more reactive than adult nodes. This makes size criteria less reliable in children. Ultrasound is the preferred first-line imaging modality because it avoids radiation and is well-tolerated.

Infectious causes dominate pediatric cervical lymphadenopathy. In Kawasaki disease, for example, ultrasound shows clusters of multiple enlarged hypoechoic nodes forming what resembles a bunch of grapes, all matted into one palpable mass. This pattern looked similar to nodes seen in acute Epstein-Barr virus infection but was distinct from bacterial lymphadenitis, where nodes tend to be more uniformly enlarged and may show frank abscess formation. In some Kawasaki patients, ultrasound was able to distinguish the pattern before other clinical features of the disease appeared.²⁴

When pediatric malignancy is a concern, the same general principles apply: round shape, absent hilum, abnormal vascularity, and necrosis all raise suspicion. But the threshold for biopsy may be different given how common benign reactive adenopathy is in children. Persistent enlargement, rapid growth, and fixation to surrounding tissue in combination with suspicious ultrasound features are what typically push clinicians toward tissue sampling.

Ultrasound-Guided Biopsy

When imaging raises concern, the next step is usually fine-needle aspiration under ultrasound guidance. The ability to watch the needle enter the node in real time improves accuracy and allows targeting of the most suspicious-looking area, such as a solid peripheral rim rather than a necrotic center where tumor cells may be sparse. For retrojugular nodes, which sit in a tricky anatomical position behind the internal jugular vein, ultrasound-guided aspiration achieved adequate diagnostic material in about 85% of cases in one series, with accuracy exceeding 95% when inadequate samples were excluded.

Targeting matters because not all parts of an abnormal node yield the same information. A biopsy of a necrotic core may come back as acellular debris, while the viable tumor tissue at the periphery is where diagnostic cells live. Ultrasound helps the clinician aim for the right zone, and that planning is informed by the very features the scan has already identified.

1
Radiography. Ultrasound of neck lymph nodes: How to do it and how do they look?
2
PubMed Central. Ultrasonographic Appearances of Cervical Lymph Nodes in Healthy Turkish Adults Subpopulation: Preliminary Study
3
PubMed Central. Correlation of Sonographic Classification of Neck Adenopathy (A-RADS) and Malignancy
4
PubMed. Ultrasonographic differentiation between metastatic and benign lymph nodes in patients with papillary thyroid carcinoma
5
PubMed Central. Ultrasound of malignant cervical lymph nodes
6
PubMed Central. Ultrasound features of extranodal extension in the metastatic cervical lymph nodes of papillary thyroid cancer: a case-control study
7
PubMed Central. Diagnostic efficacy of color Doppler ultrasound in evaluation of cervical lymphadenopathy
8
PubMed. Differential diagnosis of cervical lymphadenopathy: usefulness of color Doppler sonography
9
PubMed. Distinction between benign and malignant causes of cervical, axillary, and inguinal lymphadenopathy: value of Doppler spectral waveform analysis
10
PubMed. Usefulness of Doppler spectral analysis and power Doppler sonography in the differentiation of cervical lymphadenopathies
11
PubMed Central. Ultrasound elastography for evaluation of cervical lymph nodes
12
Ultrasound in Medicine & Biology. Quantitative Shear Wave Elastography in the Evaluation of Metastatic Cervical Lymph Nodes
13
PubMed Central. Shear Wave Elastography for Distinguishing Cervical Lymph Node Malignancy: A Prospective, Observational Study
14
PubMed. Comparison of microvascular flow imaging and contrast-enhanced ultrasound for blood flow analysis of cervical lymph node lesions
15
PubMed Central. Contrast-enhanced ultrasound for the differential diagnosis between benign and metastatic superficial lymph nodes: a meta-analysis
16
PubMed Central. Role of ultrasound in the diagnosis of cervical tuberculous lymphadenitis in children
17
Journal of Medical Ultrasound. Ultrasonographic Features of Tuberculous Cervical Lymphadenitis
18
PubMed Central. Mistakes in ultrasound diagnosis of superficial lymph nodes
19
PubMed. Sonographic evaluation of cervical lymph nodes
20
PubMed Central. A risk stratification model based on ultrasound radiologic features for cervical metastatic lymph nodes in papillary thyroid cancer
21
American Journal of Neuroradiology. A Scoring System for Prediction of Cervical Lymph Node Metastasis in Patients with Head and Neck Squamous Cell Carcinoma
22
PubMed Central. Artificial intelligence performance in ultrasound-based lymph node diagnosis: a systematic review and meta-analysis
23
PubMed Central. Ultrasound-based artificial intelligence for predicting cervical lymph node metastasis in papillary thyroid cancer: a systematic review and meta-analysis
24
Pediatrics. Ultrasonographic Evaluation of Cervical Lymph Nodes in Kawasaki Disease