Tuberculosis Lymphadenitis: Causes, Symptoms, and Treatment

Tuberculosis lymphadenitis is the most common form of tuberculosis that occurs outside the lungs, and it typically shows up as a painless, slowly growing lump in the neck. It is caused by the same bacterium responsible for pulmonary TB, but instead of attacking lung tissue, the infection settles in the lymph nodes and triggers a distinctive pattern of inflammation and tissue death. The condition is highly treatable with standard anti-TB drugs, though the road to full resolution is often slower and less predictable than with lung TB, and the diagnosis itself can be surprisingly tricky to nail down.

What Causes It

The bacterium Mycobacterium tuberculosis is responsible for the vast majority of cases. After a person inhales airborne droplets containing the bacteria, the organisms travel through the bloodstream or lymphatic channels and lodge in lymph nodes, most often in the neck. In some people this happens shortly after initial infection; in others, the bacteria sit dormant for years before reactivating. The immune system walls off the bacteria inside clusters of immune cells called granulomas, and these granulomas often develop a characteristic pattern of cheese-like tissue death known as caseous necrosis. In one multi-center laboratory study, about 85% of granulomatous tissue samples that showed this necrosis turned out to be tuberculosis, compared with only 14% of those without it.¹

A less common but important cause is Mycobacterium bovis, a closely related species found in cattle. Children who consume unpasteurized dairy products from infected herds can develop cervical lymphadenitis from M. bovis rather than M. tuberculosis.² This matters for treatment planning because M. bovis is naturally resistant to pyrazinamide, one of the standard first-line drugs.

Who Gets It

TB lymphadenitis is not evenly distributed. In high-burden regions like sub-Saharan Africa and South Asia, it accounts for a large share of all lymphadenitis diagnoses. A study in northwest Ethiopia found that roughly two-thirds of all lymphadenitis patients had the tuberculous form.³ A systematic review across Africa found that about 53% of cases were female and 68% fell in the 15-to-44 age range.⁴

In low-burden countries, the picture looks different. A study of 122 patients in Spain found that more than half were born abroad, predominantly in Pakistan, Morocco, and South America. Spanish-born patients tended to be older (median age 56 versus 32 for foreign-born patients) and had a much higher rate of HIV co-infection, at roughly a third compared with under 2% among immigrant patients.⁵ This pattern holds across Europe and North America: if you live in a wealthy country and develop TB lymphadenitis, you are statistically likely to either have been born in or traveled to a high-burden region, or to have a compromised immune system.

HIV co-infection is a major accelerant. The African systematic review found an overall HIV co-infection rate of about 52% across the continent, though with huge variation by country.⁴ The weakened immune surveillance that comes with HIV makes it far easier for TB bacteria to disseminate from the lungs to the lymph nodes and other organs.

How It Presents

The classic story is a painless, gradually enlarging lump in the neck that has been growing for one to two months by the time someone sees a doctor.⁶ The swelling usually involves a single group of cervical lymph nodes, most commonly on one side. It tends to feel firm and rubbery early on, then may soften as the tissue inside breaks down. Constitutional symptoms like fever, night sweats, and weight loss are present in some patients but far from universal, which is part of what makes this diagnosis easy to miss.

As the disease progresses, involved nodes can mat together into a larger, irregular mass. If untreated, the infection may erode through the skin, forming a draining sinus or fistula. Historically this advanced presentation was known as scrofula, a term that dates to the medieval period when surgical removal of the diseased gland was an accepted treatment.⁷ While the neck is the most common site, TB lymphadenitis can also show up in the armpit, groin, or abdominal lymph nodes.

Why Diagnosis Is Difficult

Several other conditions can look nearly identical to TB lymphadenitis. Lymphoma is one of the most important to rule out, because the clinical presentations overlap heavily: painless, persistent lymph node enlargement in a young or middle-aged adult, sometimes with systemic symptoms like fatigue and weight loss.⁸ Sarcoidosis, fungal infections, and other mycobacterial species can also produce granulomatous lymphadenopathy that mimics TB. In children, the swelling is sometimes mistaken for a dental infection or a routine reactive lymph node.⁹

When tissue samples are examined under a microscope, the presence or absence of necrosis within granulomas is one of the strongest clues. A French study of lymph node biopsies found that necrosis had about 81% sensitivity and 92% specificity for identifying TB among granulomatous lesions, meaning it catches most TB cases while rarely flagging non-TB disease as tuberculosis.¹⁰ But histology alone is not enough for a definitive diagnosis; microbiological confirmation is the goal.

Diagnostic Tools

Fine needle aspiration cytology (FNAC) is the first-line procedure in most settings. A thin needle is inserted into the swollen node to withdraw cells for examination. It is quick, inexpensive, and avoids the need for surgical biopsy in many cases. One study found FNAC had a sensitivity of about 83% and specificity of roughly 95% when compared against GeneXpert results.¹¹ The most common patterns seen on cytology are epithelioid cell granulomas with necrosis and necrosis alone.¹²

Traditional staining to look for the bacteria directly under a microscope (Ziehl-Neelsen staining) is highly specific but catches relatively few cases. One study from Australia reported a microscopy sensitivity of just 18% on fine needle aspirates.¹³ That means the vast majority of true TB cases will have a negative smear, so a negative result does not rule out TB.

The Xpert MTB/RIF test, a rapid molecular assay, has transformed the diagnostic landscape. It can detect M. tuberculosis DNA and simultaneously check for rifampicin resistance within about two hours. An early validation study showed test sensitivity of about 97% and specificity of 89% on fine-needle aspirates, and it correctly identified all six cases that were smear-negative but culture-positive.¹⁴ Larger evaluations have reported somewhat lower sensitivity, in the range of 78% to 92% depending on the reference standard used, with higher detection rates in aspirates that were purulent or caseous.¹⁵ Even so, Xpert represents a major improvement over smear microscopy for catching smear-negative cases.¹⁶

What Ultrasound Can Show

Ultrasound is often the first imaging study performed, and TB lymph nodes have several features that set them apart from reactive nodes or lymphoma. Normal or mildly inflamed lymph nodes tend to be oval-shaped with sharp borders; TB nodes are rounder, with blurred or unsharp edges due to inflammation spreading into the surrounding tissue.¹⁷ They frequently show internal areas of dark (hypoechoic) signal from cystic necrosis, and scattered bright dots from calcification are considered highly specific for TB. Another characteristic finding is a thin echogenic layer just inside the node’s outer margin.

In children, a pediatric ultrasound study found that the normal central bright stripe (the hilus) was absent or unclear in all examined TB nodes, and the blood flow pattern tended to be pushed to the edges of the node rather than entering through the center.¹⁸ Findings like abscess formation and sinus tracts can also be picked up. Contrast-enhanced CT is sometimes used to distinguish TB from lymphoma in deeper abdominal nodes, where the two can produce very different patterns of enhancement.¹⁹

Standard Drug Treatment

TB lymphadenitis is treated with the same core drugs used for pulmonary TB. The standard regimen involves an intensive phase of two months on four drugs, followed by a continuation phase of four months on fewer drugs. A typical combination for the intensive phase includes isoniazid, rifampicin, ethambutol, and pyrazinamide, taken daily; the continuation phase drops pyrazinamide and sometimes ethambutol.²⁰ Total treatment lasts six months in most guidelines, though some practitioners extend to nine months when response is slow.

One randomized trial explored whether a shorter four-month regimen containing ofloxacin (a fluoroquinolone) could replace the standard six-month course.²¹ The idea of shortening treatment is appealing because six months of daily pills is a long commitment, but evidence that shorter courses are reliable enough has been limited so far.

A challenge that surprises both patients and clinicians is how often lymph nodes fail to fully resolve by the end of treatment. A small study found that among patients who completed six months of therapy, about 63% were considered cured while roughly 37% still had residual lymphadenopathy and were extended to nine months. Even after nine months, more than half of those extended patients still had persistent swelling.²² Persistent swelling does not necessarily mean treatment failure — nodes can remain palpable for months after the bacteria are cleared — but it makes it difficult for doctors to decide when to stop or switch therapy.

Paradoxical Reactions During Treatment

One of the more unsettling experiences for patients is the paradoxical upgrading reaction, where lymph nodes actually enlarge, become inflamed, or new ones appear after treatment has already started. This happens in roughly one in five patients.⁶ It is thought to represent the immune system ramping up its attack on the bacteria as the bacterial load drops, rather than a sign that treatment is failing.

A study of 189 patients found that paradoxical reactions developed in 17%, and about three-quarters of those involved the appearance of newly inflamed glands. The reactions responded well to anti-inflammatory drugs in almost all cases, with only one patient requiring a course of steroids. No changes to the anti-TB drug regimen were needed.²³ Risk factors included having widespread bilateral lymphadenopathy at the start of treatment and having smear-positive specimens. The key practical message is that worsening lumps a few weeks into treatment do not automatically mean the drugs are not working, but the situation does need clinical evaluation to rule out other explanations.

When Surgery Is Needed

Drug therapy is always the backbone of treatment, but surgery plays a supporting role in specific situations. Current practice generally reserves surgical intervention for nodes larger than about 3 cm that are not shrinking, abscesses that need drainage, fistulas that have formed draining tracts through the skin, recurrences after completed treatment, and cases involving drug-resistant organisms.²⁴ In a Canadian series of pediatric patients, two children had culture-negative suppuration that persisted despite adequate anti-TB drugs and ultimately required surgery for cure.²⁵

Excisional biopsy, where the entire node is removed, serves both as a diagnostic and therapeutic tool. A review confirmed that while anti-bacillary treatment remains the reference standard, early surgical intervention in cases of abscess, fistula, or complications reduces the risk of disease spread and improves quality of life.²⁶ Surgery is not a replacement for drugs; patients still complete the full antibiotic course after any procedure.

Drug Resistance

Most TB lymphadenitis responds well to standard first-line drugs, but resistance is a growing concern. A molecular characterization study of 50 culture-positive lymph node isolates found that 14% carried at least one resistance-conferring gene mutation. Eight percent were resistant to isoniazid alone, and 6% were resistant to both isoniazid and rifampicin, which meets the definition of multidrug-resistant TB.²⁷ In individual cases, multidrug resistance in lymph node TB has been reported as the first sign that a patient’s previous pulmonary TB was not fully treated, requiring a switch to second-line medications.²⁸

This is one of the reasons the Xpert MTB/RIF test is so valuable: beyond confirming TB, it simultaneously flags rifampicin resistance, the single most important marker for multidrug-resistant TB. Detecting resistance early allows clinicians to avoid months on a regimen that will not work.

TB Lymphadenitis in Children

Children present a particular diagnostic challenge. They are less likely to produce the constitutional symptoms adults sometimes show, and the neck lumps can be attributed to more common childhood causes like viral infections or dental problems. A case report of a 10-year-old boy highlighted this problem: he was initially treated for cellulitis of dental origin before further investigation revealed cervical TB lymphadenitis confirmed by lymph node biopsy.⁹

In low-burden countries, the diagnosis is uncommon enough that clinicians may not consider it at first. A Canadian pediatric series found that all 16 patients presented with cervical lymphadenopathy, and 14 of the 16 were born outside Canada in TB-endemic areas. Two children in the series were found to have multidrug-resistant TB, both from prior excisional biopsies that had not been sent for culture — a reminder that any tissue removed from a suspicious lymph node should always be cultured, not just examined under a microscope.²⁵

Why Some People Get Lymph Node TB Instead of Lung TB

Exposure to M. tuberculosis does not produce the same disease in everyone. Most infections stay in the lungs, some disseminate to lymph nodes, and others reach the bones, brain, or other organs. There is growing evidence that host genetics partly determine this outcome. A genetic association study found that variants in certain genes, including LTA4H, P2RX7, and SP110, were linked to susceptibility to lymph node TB specifically, not to pulmonary TB. Pathway analysis suggested that pulmonary TB was driven more by cytokine-related genetic variants, while lymph node TB was associated with variants involved in apoptosis, the process by which cells self-destruct.²⁹

A pilot genome-wide association study identified additional genetic variants associated with extrapulmonary TB, in genes affecting cell signaling and activity.³⁰ From the immune side, extrapulmonary TB appears to involve more immune exhaustion and a stronger presence of regulatory T cells, the cells that dial down immune responses, compared with pulmonary TB.³¹ This research is still in early stages, but it helps explain a question patients often ask: why did the TB go to my lymph nodes instead of staying in my lungs? Part of the answer appears to lie in their individual immune wiring.

The Medieval Connection

TB lymphadenitis has a surprisingly long recorded history. In medieval Europe, cervical scrofula was called “The King’s Evil,” based on the belief that a royal touch could cure the swollen neck glands. French and English monarchs participated in elaborate healing ceremonies for centuries. Physicians of the era developed pathological descriptions of the characteristic tubercles and abscesses in lymph nodes, and surgical excision of the diseased glands was recommended treatment long before anyone knew a bacterium was involved.⁷ The approach has come full circle in a sense: modern guidelines still call for surgery in specific circumstances, though now paired with antibiotics that medieval surgeons could not have imagined.