Tuberculous Meningitis

Updated: Nov 10, 2021
  • Author: Gaurav Gupta, MD, FAANS, FACS; Chief Editor: Niranjan N Singh, MBBS, MD, DM, FAHS, FAANEM  more...
  • Print


Currently, more than 2 billion people (ie, one third of the world’s population) are infected with tuberculosis (TB), 10% of whom develop clinical disease, and 1.4 million of whom die of the disease annually. [50]  Tuberculous meningitis (TBM) is a manifestation of extrapulmonary TB, develping in 1%–5% of the approximately 10 million cases of TB worldwide. [47, 48, 50]  Although rare in the United States and Europe, TB is a common cause of meningitis (and the most common cause of chronic meningitis) in endemic areas worldwide, particularly among patients co-infected with HIV. [48, 49, 50]

Mycobacterium tuberculosis bacilli enter the host by droplet inhalation, after which the localized infection escalates within the lungs and then disseminates to the regional lymph nodes. The bacilli may then seed to the central nervous system (CNS) and result in three forms of CNS TB: tuberculous meningitis, intracranial tuberculoma, and spinal tuberculous arachnoiditis. [49] In the brain, the bacilli may form small subpial or subependymal foci of metastatic caseous lesions, termed Rich foci. As the disease progresses, the Rich foci enlarge and may eventually rupture into the subarachnoid space, resulting in meningitis (See Pathophysiology).

Despite great advances in immunology, microbiology, and drug development, TB remains a significant public health challenge. Although progress has been made, issues such as poverty, lack of functioning public health infrastructure, lack of funding to support basic research aimed at developing new drugs, diagnostics, and vaccines, and the co-epidemic of HIV contribute to the ongoing epidemic of TB (See Epidemiology).

If untreated, TBM may have a poor outcome and permanent neurological sequelae, thus requiring rapid diagnosis and treatment. Prognosis is related directly to the clinical stage at diagnosis (See Prognosis).

Unlike many forms of bacterial meningitis, TBM is often difficult to diagnose, as initial symptoms are generally subacute and often nonspecific (although occasionally may present more acutely), and neck stiffness is typically not present in the early course of the illness. [51, 57]  The duration of presenting symptoms may vary from 1 day to 9 months (generally, a week to a month), and the prodrome is usually nonspecific, including headache, vomiting, photophobia, and fever. Meningismus may also occur. Unlike most forms of bacterial meningitis, TBM is more likely to cause neurological deficits, including altered mental status, personality changes, and, as the lesions may result in neurovascular compression, cranial nerve deficits and infarcts. [51] (See Clinical Presentation).

The clinician should have a high index of clinical suspicion if a patient presents with a clinical picture of meningoencephalitides, especially in high-risk groups or in endemic areas. There is frequently diagnostic uncertainty when differentiating TBM from other meningoencephalitides, in particular partially treated meningitis. TBM must be differentiated not only from other forms of acute and subacute meningitis, but also from conditions such as viral infections and cerebral abscesses (See Diagnosis).

The diagnosis of TBM cannot be made or excluded solely on the basis of clinical findings. Tuberculin testing is of limited value. Variable natural history and accompanying clinical features of TBM may confuse the clinician. A lumbar puncture is necessary if meningitis is suspected, with the caveat that there is some risk of herniation of the medulla if intracranial hypertension is suspected. A small-volume lumbar puncture may be considered in such cases. CNS imaging modalities lack specificity but may aid in suggesting the diagnosis and monitoring for complications that require neurosurgical intervention (See Workup).

Prompt treatment is essential, as death or signfiicant neurological disability may occur as a result of missed diagnoses and delayed treatment. Antimicrobial therapy is best started with isoniazid, rifampin, and pyrazinamide; addition of a fourth drug is left to local choice. Although unclear, the optimal duration of antimicrobial therapy should generally be for approximately 9–12 months (longer than for isolated pulmonary TB). The benefits of adjuvant corticosteroids remain in doubt; their use in adults is controversial, though they may be indicated in the presence of increased ICP, altered consciousness, focal neurological findings, spinal block, and tuberculous encephalopathy.

In patients with evidence of obstructive hydrocephalus and neurological deterioration who are undergoing treatment for TBM, placement of a ventricular drain or ventriculoperitoneal or ventriculoatrial shunt should not be delayed. Prompt CSF diversion improves outcomes, particularly in patients presenting with minimal neurological deficit (See Treatment and Management).

New research avenues include research into vaccine design, mechanisms of drug resistance, and virulence determinants. Rapid sensitivity testing using bacteriophages considers the problem of drug resistance.

Refer to Meningitis, Meningococcal Meningitis, Staphylococcal Meningitis, Haemophilus Meningitis, Viral Meningitis, and Aseptic Meningitis for more complete information on these topics.



Many of the signs, symptoms, and sequelae of tuberculous meningitis (TBM) are the result of an immunologically directed inflammatory reaction to the infection. Mycobacterium tuberculosis bacilli enter the host by droplet inhalation, and initially infect alveolar macrophages. Localized infection worsens in the lungs, and then disseminates to the regional lymph nodes occurs, resulting in the primary complex. During this stage, a short but significant bacteremia is present that can seed bacilli to other organs. 

The bacilli may then seed to the central nervous system (CNS) and result in any of three forms of CNS TB: tuberculous meningitis, intracranial tuberculoma, and spinal tuberculous arachnoiditis. [49]  Tuberculous pneumonia may result in heavier and more prolonged tuberculous bacteremia, which renders CNS dissemination more likely, particularly if miliary TB develops. In the brain, the bacilli may form small subpial or subependymal foci of metastatic caseous lesions, known as Rich foci, after the original pathologic studies of Rich and McCordick. [1]  As the disease progresses, the Rich foci enlarge and may eventually rupture into the subarachnoid space, resulting in meningitis.

The location of the expanding tubercle (ie, Rich focus) determines the type of CNS involvement. Tubercles rupturing into the subarachnoid space cause meningitis, whereas those deeper in the brain parenchyma or in the spinal cord cause tuberculomas or abscesses. While an abscess or tuberculoma may rupture into the ventricle, a Rich focus does not.

A thick gelatinous exudate may infiltrate the cortical or meningeal blood vessels, producing inflammation, obstruction, or infarction. Unlike most forms of bacterial meningitis, TBM tends to occur at the skull base (basal meningitis), which accounts for the frequent dysfunction of cranial nerves (including III, VI, and VII), and obstructive hydrocephalus from obstruction of basilar cisterns. Subsequent neurological pathology is produced by three general processes: adhesion formation, obliterative vasculitis, and encephalitis or myelitis.

Formation of tuberculomas

Tuberculomas are conglomerate caseous foci that form within the parenchyma of the brain, as shown in the image below. They may occur anywhere in the brain, but have a predilection for subcortical structures. [52]  Centrally located, active lesions may reach considerable size without producing meningitis. [1] Under conditions of poor host resistance, this process may result in focal areas of cerebritis or frank abscess formation, but the usual course is coalescence of caseous foci and fibrous encapsulation (ie, tuberculoma).

Tuberculoma is the round gray mass in the left cor Tuberculoma is the round gray mass in the left corpus callosum. The red meninges on the right are consistent with irritation and probable meningeal reaction to tuberculosis. Courtesy of Robert Schelper, MD, Associate Professor of Pathology, State University of New York Upstate Medical University.

Tuberculomas may coalesce together or grow in size, even during ongoing antitubercular therapy [2] ; this process may have an immunological basis. [3] Tuberculomas can also involve adjacent intracranial arteries, often causing vasculitis and resulting strokes. [4] Probable embolic spread of tuberculomas in the brain in multi-drug resistant TBM has been reported. [5]

Spinal Involvement

The spinal meninges may become involved secondarily due to extension of intracranial meningitis, or may develop primarily from a tuberculous focus on the surface of the cord (which then ruptures into the subarachnoid space), or via transdural extension of an infection from caries in the spine (e.g. Pott's disease).

Pathologically, a gross granulomatous exudate fills the subarachnoid space and extends over several segments. Vasculitis involving spinal arteries and veins may occur, sometimes resulting in ischemic spinal cord infarction.

A lesion in the vertebra is almost invariably due to hematogenous spread from a pulmonary source, often involving the body of the vertebra near an intervertebral disk. Unlike bacterial osteomyelitis and discitis, in TB, involvement of the posterior elements is more common, and the disc space is more commonly spares. Associated abscesses (which may include retropharyngeal abscesses in cervical cases) also tend to be larger relative to the bony involvement. Generally, the infection begins anteriorly (inferior or superior endplate), extends under the anterior longitudinal ligament, and spreads via the venous plexus of Bateson.  

As the disease progresses, increasing decalcification and erosion result in progressive collapse of the bone and destruction of intervertebral disks, involving as many as 3-10 vertebrae in one lesion, resulting in kyphosis. The abscess may rupture intraspinally, resulting in primary spinal meningitis, hyperplastic peripachymeningitis, intraspinal abscess, or tuberculoma.

Pathological effects and sequale on vision and cerebrovascular phenomena

Papilledema is the most common sequela of TBM on the optic apparatus In children, papilledema may progress to primary optic atrophy and blindness resulting from direct involvement of the optic nerves and chiasm by basal exudates (ie, opticochiasmatic arachnoiditis). In adults, papilledema may progress more commonly to secondary optic atrophy, provided the patient survives long enough. Other causes of visual impairment include chorioretinitis, optic neuritis, internuclear ophthalmoplegia, and, occasionally, an abrupt onset of painful ophthalmoplegia.

Ocular involvement is rare in TB. When it occurs, the typical lesion is often a choroidal granuloma. Baha Ali and coworkers describe 3 cases of choroidal TB associated with 3 different clinical situations, including tuberculous meningitis, multifocal TB, and military TB with HIV. [6]

As TBM has a predilection for the skull base (unlike bacterial meningitis), cranial nerve palsies are more common. CN VI is affected most frequently by TBM, followed by CNs III, IV, VII, and, less commonly, CNs II, VIII, X, XI, and XII. [7]

Sudden onset of focal neurological deficits, including monoplegia, hemiplegia, aphasia, and tetraparesis, may occur. Most commonly, these acute deficits are due to TBM-associated vasculitis which results in ischemia (with possible hemorrhagic conversion); less commonly the etiology may be postictal, proliferative arachnoiditis, or hydrocephalus.

Vasculitis with resultant thrombosis and hemorrhagic infarction may develop in vessels that traverse the basilar exudate, spinal exudate, or lie within the brain substance. Mycobacterium also may invade the adventitia directly and initiate the process of vasculitis.

An early neutrophilic reaction is followed by infiltration of lymphocytes, plasma cells, and macrophages, leading to progressive destruction of the adventitia, disruption of elastic fibers, and, finally, intimal destruction. Eventually, fibrinoid degeneration within small arteries and veins produces aneurysms, thrombi, and focal hemorrhages, alone or in combination. [8]

Tremor is the most common movement disorder seen in the course of TBM. In a smaller percentage of patients, abnormal movements, including choreoathetosis and hemiballismus, have been observed, more so in children than in adults. In addition, myoclonus and cerebellar dysfunction have been observed. Deep vascular lesions are more common among patients with movement disorders.



Causative organism

The causative organism of tuberculous meningitis (TBM) is Mycobacterium tuberculosis. The first description of TBM is credited to Robert Whytt, in his 1768 monograph, Observations of Dropsy in the Brain. TBM first was described as a distinct pathological entity in 1836, and Robert Koch demonstrated that TB was caused by M. tuberculosis in 1882.

M. tuberculosis is an aerobic gram-positive rod that stains poorly with hematoxylin and eosin (H&E) because of its thick cell wall that contains lipids, peptidoglycans, and arabinomannans. The high lipid content in its wall makes the cells impervious to Gram staining. However, Ziehl-Neelsen stain forms a complex in the cell wall that prevents decolorization by acid or alcohol, and the bacilli are stained a bright red, which stands out clearly against a blue background.

Mycobacteria vary in appearance from spherical to short filaments, which may be branched. Although they appear as short to moderately long rods, they can be curved and frequently are seen in clumps. Individual bacilli generally are 0.5–1 µm in diameter and 1.5–10 µm long. They are nonmotile and do not form spores.

One of the distinct characteristics of mycobacteria is their ability to retain dyes within the bacilli that usually are removed from other microorganisms by alcohols and dilute solutions of strong mineral acids such as hydrochloric acid. This ability is attributed to a waxlike layer composed of long-chain fatty acids, the mycolic acids, in their cell wall. As a result, mycobacteria are termed acid-fast bacilli.

The mechanisms by which neurovirulence may occur are unknown.

Risk factors

Human migration plays a large role in the epidemiology of TB. Massive human displacement during wars and famines has resulted in increased case rates of TB and an altered geographic distribution. With the advent of air travel, TB has a global presence. In the United States, the prevalence of TB, mostly in foreign-born persons, has steadily increased.

Once infected with M. tuberculosis, HIV co-infection is the strongest risk factor for progression to active TB; the risk has been estimated to be as great as 10% per year, compared with 5-10% lifetime risk among persons with TB but not HIV infection. Although patients who have HIV infection and TB are at increased risk for TBM, the clinical features and outcomes of TB do not seem to be altered by HIV. Go to HIV-1 Associated CNS Conditions - Meningitis for more complete information on this topic.

Patients infected with HIV, especially those with AIDS, are at very high risk of developing active TB when exposed to a person with infectious drug-susceptible or drug-resistant TB. They have a higher incidence of drug-resistant TB, in part due to Mycobacterium avium-intracellulare, and have worse outcomes.

Other predisposing factors for the development of active TB include malnutrition, alcoholism, substance abuse, diabetes mellitus, corticosteroid use, malignancy, and head trauma. Homeless persons, people in correctional facilities, and residents of long-term care facilities also have a higher risk of developing active TB compared with the general population.



More than 2 billion people (ie, one third of the world’s population) are infected with tuberculosis (TB), 10% of whom  develop clinical disease, and 1.1–1.3 million of whom die of the disease annually. [50]  Tuberculous meningitis (TBM) is a manifestation of extrapulmonary tuberculosis, develping in 1%–5% of the approximately 10 million cases of TB worldwide. [47, 48, 50]  Although rare in the United States and Europe, TB is a common cause of meningitis (and most common cause of chronic meningitis) in endemic areas worldwide, particularly among patients co-infected with HIV. [48, 49, 50]  TB is the seventh leading cause of death and disability worldwide. 

United States statistics

TB in the United States is uncommon; in 2019 there were 8,916 confirmed cases of TB (2.7 cases per 100,000 persons), downtrending from 11,077 approximately a decade earlier. [53]  The majority of cases are among foreign-born individuals. Of confirmed TB cases in 2019, there were 1,833 reported cases of extrapulmonary TB, 85 of whom had TBM. [53]  These figures have also been downtrending over the past decade from 2,411 and 138 cases in 2010, respectively. In 2018, of the 9,024 patients diagnosed with TB in the United States, 542 died of the disease (6.0%).

Between 1969 and 1973, TBM accounted for approximately 4.5% of the total extrapulmonary TB morbidity in the United States. Between 1975 and 1990, 3,083 cases of TBM were reported by the US Centers for Disease Control and Prevention (CDC), an average of 193 cases per year, accounting for 4.7% of total extrapulmonary TB cases during that 16-year period. In 1990, however, 284 cases of TBM were reported, constituting 6.2% of the morbidity attributed to extrapulmonary TB. This increase in TBM was most likely due to increasing CNS TB among patients with HIV/AIDS and to the increasing incidence of TB among infants, children, and young adults of minority populations. Data suggest that TBM accounts for 2.1% of pediatric cases and 9.1% of extrapulmonary TB cases. [9] TB accounts for approximately 0.04% of all cases of chronic suppurative otitis media. [10] The Tuberculosis: Advocacy Report released by the World Health organization (WHO) in 2003 suggests the persistence of TB otitis, as well as possibly an increase in the incidence of TB otitis. [11] Tuberculomas account for 10%–30% of intracranial masses in TB-endemic areas.

International statistics

The WHO estimates that the incidence of new TB infections has been increasing over the past decade, from 5.7 million diagnosed in 2009 to 7.1 million in 2019. [50]  More than a quarter of total new cases annually (estimated to be approximately 10 million) may be missed. From 2013 to 2019, the countries with the greatest contributions to new diagnoses were India and Indonesia. In 2019, there were approximately 1.2 million TB deaths among HIV-negative inidividuals (a reduction from 1.7 million in 2000), and an additional 208,000 among HIV-positive patients. TBM develops in 1%–5% of the approximately 10 million cases of TB worldwide. [47, 48, 50]

In 2005, the TB incidence rate was stable or in decline in all six WHO regions. However, the total number of new TB cases was still rising slowly; the case-load continues to grow in the African, eastern Mediterranean, and Southeast Asia regions. [13] In many areas of Africa and Asia, the annual incidence of TB infection for all ages is approximately 2%, which would yield an estimated 200 cases of TB per 10,000 population per year. Approximately 15%–20% of these cases occur in children younger than 15 years.

The worldwide prevalence of TB in children is difficult to assess because data are scarce and poorly organized, although the WHO estimates that approximately 10% of cases occur in children, primarily among those ages 2–4 years. [47] The available reports likely underestimate the true incidence, however, as lack of surveillance testing in most areas of the world limits the ability to assess the true prevalence of the disease. The developing world has 1.3 million cases of TB and 40,000 TB-related deaths annually among children younger than 15 years. In the developing world, 10%–20% of persons who die of TB are children. TBM complicates approximately 1 of every 300 untreated primary TB infections.

Age distribution for TBM

In the United States in 2019, 4.1% of cases of TB were diagnosed in children aged 14 years and younger, 9.5% of cases were amont those 15–24 years of age, 29.3% in those 25–44 years of age, 29.9% in those 45–64 years of age, and 27.2% in those >/= 65 years of age. [53]  Among TB patients, the highest rates of TBM proportionally per TB infection are found in children aged 2–4 years. [47] . A study of TBM in Texas during the period 2010–2017 found that although children 4 years of age and younger accounted for 4.0% of TB cases, they accounted for 14.6% of cases of TBM. [54]  TBM is uncommon, however, in children younger than 6 months and extremely rare in infants younger than 3 months, as the causative pathological sequelae generally take at least 3 months to develop.

Children aged 5–14 years often have been referred to as the favored age because they have lower rates of TB than any other age group. Younger children are more likely to develop meningeal, disseminated, or lymphatic TB, whereas adolescents more frequently present with pleural, genitourinary, or peritoneal disease. Childhood TB has a limited influence on the immediate epidemiology of the disease because children rarely are a source of infection to others.

Sex distribution for TBM

Among persons younger than 20 years, TB infection rates are similar for both sexes; the lowest rates are observed in children aged 5–14 years. During adulthood, TB infection rates are consistently higher for men than for women; the male-to-female ratio is approximately 2:1. [53]

Prevalence of TBM by race

In the United States, 88% of TB cases occur among racial and ethnic minorities. [53] Case rates are lowest among Caucasians, and highest among Asians and Pacific Islanders. Rates among African Americans and Hispanics are intermediate. 

Several important factors likely contribute to the disproportionate burden of TB in minorities. In foreign-born persons from countries where TB is common, active TB disease may result from infection acquired in the country of origin. Approximately 95% of cases in the Asian/Pacific Islander group occurred in foreign-born persons, compared with 70% of cases in Hispanics and 20% of cases in non-Hispanic blacks.

In racial and ethnic minorities, unequal distribution of TB risk factors, such as HIV infection, also may contribute to an increased exposure to TB or to the risk of developing active TB once infected with M. tuberculosis. However, much of the increased risk of TB in minorities has been linked to lower socioeconomic status and the effects of crowding, particularly among US-born persons.




Overall, tuberculosis meningitis (TBM) results in death or significant disability in approximately half of cases, and mortality rates among patients co-infected with HIV is apprixiamtely 50%. [55, 57] In a long-term longitudinal cohort study of patients with TBM in New York City from 1992 to 2001, [56]  376 patients had TBM, the majority of whom (63%) were co-infected with HIV. Ten percent of patients died prior to initiating antituberculosis therapy (median 7 days from diagnosis), and 57% died prior to completing the treatment. Overall, long-term survival was approximately 40%; however, among HIV-negative patients with TBM, this figure was approximately 65%. Antibiotic-resistance correlated with reduced survival, and multi-drug resistant (MDR) TBM was nearly universally fatal (95% case fatality rate). 

Prediction of outcome

Prediction of prognosis of TBM is difficult because of the protracted course, diversity of underlying pathological mechanisms, variation of host immunity, and virulence of M. tuberculosis. Initially, only clinical indices were used for predicting the outcome, such as level of consciousness, stage of meningitis, bacillus Calmette-Guérin (BCG) vaccination status, cerebrospinal fluid (CSF) findings, and evidence of raised intracranial pressure (ICP). After MRI and CT scanning became available, radiological findings such as hydrocephalus, infarction, severity of exudate, and tuberculoma also were considered for predicting the prognosis of TBM.

Prognosis is in part related to the clinical stage of TBM and neurological status at diagnosis, as early diagnosis and treatment is the strongest determinant of favorable outcome. [57]  A widely accepted grade of the disease, which factors in the patient's neurological status via the modified Glasgow Coma Scale, is the British Medical Research Council TBM grade, a strong predictor of outcome: Grade 1 GCS = 15, no focal deficits; Grade 2 GCS = 15 with focal deficits or GCS = 11–14, or Grade 3 GCS of < /= 10. [58]  In general, patients with poorer neurological exam on presentation have poorer outcomes. 

Co-infection with HIV, particularly ineffectively treated HIV, is also a strong predictor of poor outcome. [56, 58]  

Resistance to antitubercular therapy (particularly rifampin and isoniazid) is a strong predictor of poor outcome, as was shown in the New York study. [56]

In a study that looked at clinical parameters, laboratory studies, and CT scan features in 49 adults and children with TBM used a multivariate logistic regression model to show that the most significant variables for predicting outcome in TBM were age, stage of disease, focal weakness, cranial nerve palsy, and hydrocephalus. [15] Children with advanced disease with neurological complications have poor outcomes.

The occurrence of syndrome of inappropriate diuretic hormone secretion (SIADH) as well as Cerebral Salt Wasting occurs in > 50% of TBM patients, and may suggest a poorer prognosis. [59]

Visual disturbances may suggest a poorer outcome, as they are often the result of associated optochiasmatic arachnoiditis or optochiasmal tuberculoma. [16]  

Seizures affect 16.3%–31.5% of TBM patients, with higher rates in children and in those with HIV, and worsen mortality and disability. [60]  Surviving patients who had TBM-associated seizures may develop chronic epilepsy. Unlike many other forms of meningitis, there are more EEG abnormalities in TBM. [60]  Motor-evoked potentials and somatosensory evoked potentials also have been reported to predict a 3-month outcome of TBM. Misra et al found that focal weakness, Glasgow Coma Scale score, and somatosensory evoked potential findings were the best predictors of 6-month outcome in patients with TBM. [17]

Kumar et al reported that children with TBM who have been vaccinated with BCG appear to maintain better mentation and have superior outcomes. They believe this may be explained, in part, by the better immune response to infection, as is reflected in the higher CSF cell counts in their patient group. [18]


Patient Education

Health education efforts must be directed at the patients to make them more informed and aware of all aspects of the disease and its treatment. Patients must be informed of the basic rules to prevent spreading the infection to others in the family or the community.

Whereas one end of the spectrum of educational efforts is directed toward the health-related behavior of the general public, the other end should be directed toward gaining the support of those who influence health policies and funding of governments and institutions. Information, education, and communication (IEC) campaigns should be designed to act as an intermediary between the 2 groups. This strategy includes social marketing, health promotion, social mobilization, and advocacy programs.

For patient education resources, see the Bacterial and Viral Infections Center, Brain and Nervous System Center, and Procedures Center, as well as Tuberculosis, Meningitis in Adults, Meningitis in Children, and Spinal Tap.