Tuberculosis disease is caused by infection with organisms of the Mycobacterium tuberculosis complex, which includes M tuberculosis, Mycobacterium bovis , and Mycobacterium africanum. M africanum is rare in the United States, and clinical laboratories do not distinguish it routinely. M bovis can be distinguished routinely, and although the spectrum of illness that is caused by M bovis is similar to that of M tuberculosis , the Epidemiology, Treatment, and prevention are distinct. Most infections caused by M tuberculosis complex in children and adolescents are asymptomatic. When tuberculosis disease does occur, Clinical Manifestations most often appear 1 to 6 months after infection and include fever, weight loss, or poor weight gain and possibly growth delay, cough, night sweats, and chills. Chest radiographic findings after infection range from normal to diverse abnormalities, such as lymphadenopathy of the hilar, subcarinal, paratracheal, or mediastinal nodes; atelectasis or infiltrate of a segment or lobe; pleural effusion; cavitary lesions; or miliary disease. Extrapulmonary manifestations include meningitis and granulomatous inflammation of the lymph nodes, bones, joints, skin, and middle ear and mastoid. Renal tuberculosis and progression to disease from latent tuberculosis infection ("adult-type pulmonary tuberculosis") are unusual in younger children but can occur in adolescents. In addition, chronic abdominal pain with intermittent partial intestinal obstruction can be present in disease caused by M bovis . Clinical findings in patients with drug-resistant tuberculosis disease are indistinguishable from manifestations in patients with drug-susceptible disease.

The agent is M tuberculosis complex, a group of closely related acid-fast bacilli (AFB): M tuberculosis, M bovis , and M africanum .

• Positive tuberculin skin test (TST) result . A positive TST result (see Table 3.79) indicates possible infection with M tuberculosis complex. Tuberculin reactivity appears 2 to 10 weeks after initial infection; the median interval is 3 to 4 weeks (see Tuberculin Testing).


• Positive interferon-gamma release assay (IGRA) . A positive IGRA result indicates possible infection with M tuberculosis complex.


• Exposed person refers to a person who has had recent contact with another person with suspected or confirmed contagious pulmonary tuberculosis disease and who has a negative TST or IGRA result, normal physical examination findings, and chest radiographic findings that are not compatible with tuberculosis. Some exposed people become infected (and subsequently, most have a positive TST or IGRA result) and some people do not become infected after exposure; the 2 groups cannot be distinguished initially.


• Source case is defined as the person who has transmitted infection with M tuberculosis complex to another person who subsequently has either latent tuberculosis infection or tuberculosis disease.


• Latent tuberculosis infection (LTBI) is defined as M tuberculosis complex infection in a person who has a positive TST or IGRA result, no physical findings of disease, and chest radiograph findings that are normal or reveal evidence of healed infection (eg, calcification in the lung, hilar lymph nodes, or both).


• Tuberculosis disease is defined as disease in a person with infection in whom symptoms, signs, or radiographic manifestations caused by M tuberculosis complex are apparent; disease may be pulmonary, extrapulmonary, or both. Infectious tuberculosis refers to tuberculosis disease of the lungs or larynx in a person who has the potential to transmit the infection to other people.


• Directly observed therapy (DOT) is defined as an intervention by which medication is administered directly to the patient by a health care professional or trained third party (not a relative or friend), who observes and documents that the patient ingests each dose of medication.


• Multiply drug-resistant (MDR) tuberculosis is defined as tuberculosis infection or disease caused by a strain of M tuberculosis complex that is resistant to at least isoniazid and rifampin, the 2 first-line drugs with greatest efficacy.


• Extensively drug-resistant (XDR) tuberculosis is a subset of MDR tuberculosis. It is defined as infection or disease caused by a strain of M tuberculosis complex that is resistant to isoniazid and rifampin, at least 1 fluoroquinolone, and at least 1 of the following parenteral drugs: amikacin, kanamycin, or capreomycin.


• Bacille Calmette-Guérin (BCG) is an attenuated vaccine strain of M bovis . BCG rarely is administered to children in the United States but is probably the most widely used vaccine in the world. An isolate of BCG can be distinguished from wild-type M bovis only in a reference laboratory.

Table 3-79

Case rates of tuberculosis for all ages are higher in urban, low-income areas and in nonwhite racial and ethnic groups; two thirds of reported cases in the United States occur in nonwhite people. In recent years, foreign-born children have accounted for more than one quarter of newly diagnosed cases in children 14 years of age or younger. Specific groups with greater LTBI and disease rates include immigrants, international adoptees, and refugees from or travelers to high-prevalence regions (eg, Asia, Africa, Latin America, and countries of the former Soviet Union); homeless people; and residents of correctional facilities.

Infants and postpubertal adolescents are at increased risk of progression of LTBI to tuberculosis disease. Other predictive factors for development of disease include recent infection (within the past 2 years); immunodeficiency, especially from human immunodeficiency virus (HIV) infection; use of immunosuppressive drugs, such as prolonged or high-dose corticosteroid therapy or chemotherapy; intravenous drug use; and certain diseases or medical conditions, including Hodgkin disease, lymphoma, diabetes mellitus, chronic renal failure, and malnutrition. There have been reports of tuberculosis disease in adolescents and adults being treated for arthritis, inflammatory bowel disease, and other conditions with tumor necrosis factor alpha (TNF-alpha) antagonists, such as infliximab and etanercept. Before use of TNF-alpha antagonists, patients should be screened for risk factors for M tuberculosis complex infection and have a TST or IGRA performed.

A diagnosis of LTBI or tuberculosis disease in a young child is a public health sentinel event usually representing recent transmission . Transmission of M tuberculosis complex is airborne, with inhalation of droplet nuclei usually produced by an adult or adolescent with contagious pulmonary or laryngeal tuberculosis disease. M bovis is transmitted most often by unpasteurized dairy products. The duration of contagiousness of an adult receiving effective Treatment depends on drug susceptibilities of the organism, the number of organisms in sputum, and frequency of cough. Although contagiousness usually lasts only a few days to weeks after initiation of effective drug therapy, it can last longer, especially when the adult patient has cavitary disease, does not adhere to medical therapy, or is infected with a drug-resistant strain. If the sputum smear is negative for AFB organisms on 3 separate days and the patient has improved clinically, the treated person can be considered at low risk of disease transmission. Children younger than 10 years of age with pulmonary tuberculosis rarely are contagious, because their pulmonary lesions are small (paucibacillary disease), cough is not productive, and few or no bacilli are expulsed. Unusual cases of adult-form pulmonary disease in young children can be highly contagious.

The incubation period from infection to development of a positive TST or IGRA result is 2 to 10 weeks. The risk of developing tuberculosis disease is highest during the 6 months after infection and remains high for 2 years; however, many years can elapse between initial tuberculosis infection and tuberculosis disease.

Isolation of M tuberculosis complex by culture from specimens of gastric aspirates, sputum, bronchial washings, pleural fluid, cerebrospinal fluid (CSF), urine, or other body fluids or a biopsy specimen establishes the diagnosis. Children older than 5 years of age and adolescents frequently can produce sputum by induction with aerosolized hypertonic saline. Studies have demonstrated successful collections of induced sputum from infants with pulmonary tuberculosis, but this requires special expertise. The best specimen for diagnosis of pulmonary tuberculosis in any child or adolescent in whom the cough is nonproductive or absent and sputum cannot be induced is an early morning gastric aspirate. Gastric aspirate specimens should be obtained with a nasogastric tube on awakening the child and before ambulation or feeding.¹ Aspirates collected on 3 separate days should be submitted for testing. Results of AFB smears of gastric aspirates usually are negative, and false-positive smear results caused by the presence of nontuberculous mycobacteria can occur. Gastric aspirates have the highest culture yield in young children on the first day of collection. Fluorescent staining methods for gastric aspirate smears are more sensitive and, if available, are preferred. The overall diagnostic yield of gastric aspirates is less than 50%. Histologic examination for and demonstration of AFB and granulomas in biopsy specimens from lymph node, pleura, mesentery, liver, bone marrow, or other tissues can be useful, but M tuberculosis complex organisms cannot be distinguished reliably from other mycobacteria in stained specimens. Regardless of results of the AFB smears, each specimen should be cultured.

Because M tuberculosis complex organisms are slow growing, detection of these organisms may take as long as 10 weeks using solid media; use of liquid media allows detection within 1 to 6 weeks and usually within 3 weeks. Even with optimal culture techniques, M tuberculosis complex organisms are isolated from fewer than 50% of children and 75% of infants with pulmonary tuberculosis diagnosed by other clinical criteria. Species identification of isolates from culture can be more rapid if a DNA probe or high pressure liquid chromatography is used. The differentiation between M tuberculosis and M bovis usually is based on pyrazinamide resistance, which is characteristic of almost all M bovis isolates.

One nucleic acid amplification test for rapid diagnosis is licensed by the Food and Drug Administration (FDA) for acid-fast stain positive respiratory tract specimens only, and another one is approved for any respiratory tract specimens, but these tests have decreased sensitivity for gastric aspirate, CSF, and tissue specimens, with false-negative and false-positive results reported.

Identification of the culture-positive source case supports the child's presumptive diagnosis and provides the likely drug susceptibility of the child's organism. Culture material should be collected from children with evidence of tuberculosis disease, especially when (1) an isolate from a source case is not available; (2) the presumed source case has drug-resistant tuberculosis; (3) the child is immunocompromised (eg, HIV infection); or (4) the child has extrapulmonary disease. Drug resistance cannot be confirmed without a bacterial isolate.

Tuberculin Testing

The TST is the most common method for diagnosing LTBI in asymptomatic people. The Mantoux method consists of 5 tuberculin units of purified protein derivative (0.1 mL) injected intradermally using a 27-gauge needle and a 1.0-mL syringe into the volar aspect of the forearm. Creation of a palpable induration 6 to 10 mm in diameter is crucial to accurate testing. Multiple puncture tests are not recommended, because they lack adequate sensitivity and specificity.

A TST should be administered to children who are at increased risk of acquiring LTBI and tuberculosis disease (see Table 3.80). Routine TST administration, including programs based at schools, child care centers, and camps that include populations at low risk, is discouraged, because it results in either a low yield of positive results or a large proportion of false-positive results, leading to an inefficient use of health care resources. Simple questionnaires can identify children with risk factors for LTBI who then should be tested with a TST (see Table 3.81). Risk assessment for tuberculosis should be performed at first contact with a child and every 6 months thereafter for the first 2 years of life (eg, 2 weeks and 6, 12, 18, and 24 months of age). If at any time, tuberculosis disease is suspected, a TST should be performed, although a negative result should be considered as especially unreliable in infants younger than 3 months of age. After 2 years of age, risk assessment for tuberculosis should be performed annually, if possible.

Table 3-80

Table 3-81

Recommendations for use of the TST are independent of those for immunization. Tuberculin testing at any age is not required before administration of live-virus vaccines. A TST can be applied at the same visit during which these vaccines are administered. Measles vaccine temporarily can suppress tuberculin reactivity for at least 4 to 6 weeks. The effect of live-virus varicella, yellow fever, and live-attenuated influenza vaccines on TST reactivity and IGRA results is not known. In the absence of data, the same TST spacing recommendation should be applied to these vaccines as described for MMR. There is no evidence that inactivated vaccines, polysaccharide vaccines, or recombinant or subunit vaccines or toxoids interfere with immune response to TST.

Administration of TSTs and interpretation of results should be performed by experienced health care professionals who have been trained in the proper methods, because administration and interpretation by unskilled people and family members are unreliable. The recommended time for assessing the TST result is 48 to 72 hours after administration. However, induration that develops at the site of administration more than 72 hours later should be measured, and some experts advise that this should be considered the result. The diameter of induration in millimeters is measured transversely to the long axis of the forearm. Positive test results, as defined in Table 3.79, can persist for several weeks.

A negative TST result does not exclude LTBI or tuberculosis disease. Approximately 10% to 40% of immunocompetent children with culture-documented tuberculosis disease do not react initially to a TST. Host factors, such as young age, poor nutrition, immunosuppression, other viral infections (especially measles, varicella, and influenza), recent tuberculosis infection, and disseminated tuberculosis disease can decrease TST reactivity. Many children and adults coinfected with HIV and M tuberculosis complex do not react to a TST. Control skin tests to assess cutaneous anergy are not recommended routinely.

Interpretation of TST Results (see Table 3.79)

Classification of TST results is based on epidemiologic and clinical factors. The size of induration (mm) for a positive result varies with the person's risk of LTBI and progression to tuberculosis disease.

Current guidelines from the Centers for Disease Control and Prevention (CDC), American Thoracic Society, and American Academy of Pediatrics accept 15 mm or greater of induration as a positive TST result for any person. Interpretation of 5 mm or more or 10 mm or more induration is summarized in Table 3.79 . Interpretation is aided by knowledge of the child's risk factors for LTBI and tuberculosis disease. Prompt clinical and radiographic evaluation of all children and adolescents with a positive TST reaction is recommended.

Interpretation of TST Results in Previous Recipients of BCG Vaccine

Generally, interpretation of TST results in BCG recipients is the same as for people who have not received BCG vaccine. After BCG immunization, distinguishing between a positive TST result caused by pathogenic M tuberculosis complex infection and that caused by BCG can be impossible. Reactivity of the TST after receipt of BCG vaccine does not occur in some patients. The size of the TST reaction (ie, mm of induration) attributable to BCG immunization depends on many factors, including age at BCG immunization, quality and strain of BCG vaccine used, number of doses of BCG vaccine received, nutritional and immunologic status of the vaccine recipient, frequency of TST administration, and time lapse between immunization and TST.

Tuberculosis disease should be suspected strongly in any symptomatic person regardless of a TST result and history of BCG immunization. When evaluating an asymptomatic child who has a positive TST result and who possibly received BCG vaccine, certain factors, such as documented receipt of multiple BCG immunizations (as evidenced by BCG scars), decrease the likelihood that the positive TST result is attributable to LTBI. Evidence that increases the probability that a positive TST result is attributable to LTBI includes known contact with a person with contagious tuberculosis, a family history of tuberculosis disease, a long interval (more than 5 years) since neonatal BCG immunization, and a TST reaction 15 mm or greater.

Prompt clinical and radiographic evaluation of all children with a positive TST reaction is recommended. Chest radiographic findings of a granuloma, calcification, or adenopathy can be caused by infection with M tuberculosis complex but not by BCG immunization. BCG can cause suppurative lymphadenitis in the regional lymph node drainage of the infectious site of a healthy child and can cause disseminated disease in children with some forms of immunodeficiency.

Recommendations for TST Use

The most reliable strategies for preventing LTBI and tuberculosis disease in children are based on thorough and expedient contact investigations rather than nonselective skin testing of large populations. Contact investigations are public-health interventions that should be coordinated through the local public health department. Specific recommendations for TST use are given in Table 3.80 . All children need routine health care evaluations that include an assessment of their risk of exposure to tuberculosis. Only children deemed to have increased risk of contact with people with contagious tuberculosis or children with suspected tuberculosis disease should be considered for a TST. Household investigation is indicated whenever a TST result of a household member converts from negative to positive (indicating recent infection).

Immunologic-Based Testing

QuantiFERON-TB Gold and Gold In-Tube and T-SPOT.TB are IGRAs. These FDA-approved tests measure ex vivo interferon-gamma production from T lymphocytes in response to stimulation with antigens that are fairly specific to M tuberculosis complex. As with TSTs, IGRAs cannot distinguish between latent infection and disease, and a negative result from these tests cannot exclude the possibility of tuberculosis infection or disease in a patient with findings that raise suspicion for these conditions. The sensitivity of these blood tests is similar to that of TSTs for detecting infection in adults and children who have untreated culture-confirmed tuberculosis. The specificity of IGRAs is higher than that for TSTs, because the antigens used are not found in BCG or most pathogenic nontuberculous mycobacteria (eg, are not found in M avium complex but are found in M kansasii, M fortuitum , and M marinum ). IGRAs are recommended by the Centers for Disease Control and Prevention, and some experts prefer IGRAs for use in adults in all circumstances in which a TST is used. The published experience with testing children with IGRAs is less extensive than for adults, but a number of studies have demonstrated that IGRAs perform well in most children 4 years of age and older. Some children who received BCG vaccine may have a false-positive TST result, and LTBI is overestimated by use of the TST, even in these circumstances. However, the correct interpretation of a negative IGRA test result in a child with a positive TST result remains challenging because of the current absence of longitudinal studies to determine the negative predictive value of the IGRAs (when the TST result is positive and the IGRA result is negative).

At this time, neither an IGRA nor the TST can be considered a "gold standard" for diagnosis of LTBI. Current recommendations for use of IGRAs in children are as follows:

• For immune-competent children 5 years of age and older, IGRAs can be used in place of a TST to confirm cases of tuberculosis or cases of LTBI and likely will yield fewer false-positive test results.


• Children with a positive result from an IGRA should be considered infected with M tuberculosis complex. A negative IGRA result cannot universally be interpreted as absence of infection.


• Because of their higher specificity and lack of cross-reaction with BCG, IGRAs may be useful in children who have received BCG vaccine. IGRAs may be useful to determine whether a BCG-immunized child with a reactive TST more likely has LTBI or has a false-positive TST reaction caused by the BCG.


• IGRAs cannot be recommended routinely for use in children younger than 5 years of age or for immune-compromised children of any age because of a lack of published data about their utility with these groups.


• Indeterminate IGRA results do not exclude tuberculosis infection and should not be used to make clinical decisions.

Serologic tests for tuberculosis disease that are used in some Asian and African countries have unsatisfactory sensitivity and specificity, and none of them have been approved for use in the United States.
HIV Infection

Children with HIV infection are considered at high risk of tuberculosis, and an annual TST beginning at 3 through 12 months of age is recommended. Children who have tuberculosis disease should be tested for HIV infection.

Table 3-82

Specific Drugs

Antituberculosis drugs kill M tuberculosis complex organisms or inhibit multiplication of the organism, thereby arresting progression of LTBI and preventing most complications of early tuberculosis disease. Chemotherapy does not cause rapid disappearance of already caseous or granulomatous lesions (eg, mediastinal lymphadenitis). Dosage recommendations and the more commonly reported adverse reactions of major antituberculosis drugs are summarized in Tables 3.82 and 3.83 . For Treatment of tuberculosis disease, these drugs always must be used in recommended combination to minimize emergence of drug-resistant strains. Use of nonstandard regimens for any reason (eg, drug allergy or drug resistance) should be undertaken only in consultation with an expert in treating tuberculosis.

Table 3-83

Isoniazid is bactericidal, rapidly absorbed, and well tolerated and penetrates into body fluids, including cerebrospinal fluid (CSF). Isoniazid is metabolized in the liver and excreted primarily through the kidneys. Hepatotoxic effects are rare in children but can be life threatening. In children and adolescents given recommended doses, peripheral neuritis or seizures caused by inhibition of pyridoxine metabolism are rare, and most do not need pyridoxine supplements. Pyridoxine supplementation is recommended for exclusively breastfed infants and for children and adolescents on meat- and milk-deficient diets; children with nutritional deficiencies, including all symptomatic HIV-infected children; and pregnant adolescents and women. For infants and young children, isoniazid tablets can be pulverized.

Rifampin is a bactericidal agent in the rifamycin class of drugs that is absorbed rapidly and penetrates into body fluids, including CSF. Other drugs in this class approved for treating tuberculosis are rifabutin and rifapentine. Rifampin is metabolized by the liver and can alter the pharmacokinetics and serum concentrations of many other drugs. Rare adverse effects include hepatotoxicity, influenza-like symptoms, and pruritus. Rifampin is excreted in bile and urine and can cause orange urine, sweat, and tears and discoloration of soft contact lenses. Rifampin can make oral contraceptives ineffective, so other birth control methods should be adopted when rifampin is administered to sexually active female adolescents and adults. For infants and young children, the contents of the capsules can be suspended in wild cherry-flavored syrup or sprinkled on semisoft foods (eg, applesauce). M tuberculosis complex isolates that are resistant to rifampin are uncommon in the United States. Rifabutin is a suitable alternative to rifampin in children with HIV infection receiving highly active antiretroviral therapy that proscribes the use of rifampin; however, experience in children is limited. Major toxicities of rifabutin include leukopenia, gastrointestinal tract upset, polyarthralgia, rash, increased transaminase concentrations, and skin and secretion discoloration (pseudojaundice). Anterior uveitis has been reported among children receiving rifabutin as prophylaxis or as part of a combination regimen for Treatment, usually when administered at high doses. Rifabutin also increases hepatic metabolism of many drugs but is a less potent inducer of cytochrome P450 enzymes than rifampin and has fewer problematic drug interactions than rifampin. However, adjustments in dose of rifabutin and the coadministered antiretroviral drugs may be necessary for certain combinations. Rifapentine is a long-acting rifamycin that permits weekly dosing in select adults, but it has not been evaluated in pediatric patients.

Pyrazinamide attains therapeutic CSF concentrations, is detectable in macrophages, is administered orally, and is metabolized by the liver. Administration of pyrazinamide with isoniazid and rifampin allows for 6-month regimens in patients with drug-susceptible tuberculosis. Almost all isolates of M bovis are resistant to pyrazinamide, precluding 6-month therapy for this pathogen. In daily doses of 40 mg/kg per day or less, pyrazinamide seldom has hepatotoxic effects and is well tolerated by children. Some adolescents and many adults develop arthralgia and hyperuricemia because of inhibition of uric acid excretion. Pyrazinamide must be used with caution in people with underlying liver disease. Pyrazinamide, when administered with rifampin, is associated with high rates of hepatotoxicity.

Ethambutol is well absorbed after oral administration, diffuses well into tissues, and is excreted in urine. However, concentrations in the CSF are low. At 20 mg/kg per day, ethambutol is bacteriostatic, and its primary therapeutic role is to prevent emergence of drug resistance. Ethambutol can cause reversible or irreversible optic neuritis, but reports in children with normal renal function are rare. Children who are receiving ethambutol should be monitored monthly for visual acuity and red-green color discrimination if they are old enough to cooperate. Use of ethambutol in young children whose visual acuity cannot be monitored requires consideration of risks and benefits.

Streptomycin is regarded as a "second-line" drug and is available only on a limited basis. It is administered intramuscularly. When streptomycin is not available, kanamycin, amikacin, or capreomycin are alternatives that can be prescribed for the initial 4 to 8 weeks of therapy. Patients who receive any of these drugs should be monitored for otic, vestibular, and renal toxicity.

The less commonly used (eg, "second-line") antituberculosis drugs, their doses, and adverse effects are listed in Table 3.84 . These drugs have limited usefulness because of decreased effectiveness and greater toxicity and should be used only in consultation with a specialist familiar with childhood tuberculosis. Ethionamide is an orally administered antituberculosis drug that is well tolerated by children, achieves therapeutic CSF concentrations, and may be useful for Treatment of people with meningitis or drug-resistant tuberculosis. Fluoroquinolones have antituberculosis activity and can be used in special circumstances. Because some fluoroquinolones are approved by the FDA for use only in people 18 years of age and older, their use in younger patients necessitates careful assessment of the potential risks and benefits (see Antimicrobial Agents and Related Therapy).

Table 3-84

Occasionally, a patient cannot tolerate oral medications. Isoniazid, rifampin, streptomycin and related drugs, and fluoroquinolones can be administered parenterally.

Therapy for LTBI

Isoniazid given to adults who have LTBI (ie, no clinical or radiographic abnormalities suggesting tuberculosis disease) provides substantial protection (54%-88%) against development of tuberculosis disease for at least 20 years. Among children, efficacy approaches 100% with appropriate adherence to therapy. All infants, children, and adolescents who have a positive TST result but no evidence of tuberculosis disease and who never have received antituberculosis therapy should be considered for isoniazid unless resistance to isoniazid is suspected (ie, known exposure to a person with isoniazid-resistant tuberculosis) or a specific contraindication exists. Isoniazid, in this circumstance, is therapeutic and prevents development of disease. A physical examination and chest radiograph should be obtained at the time isoniazid therapy is initiated to exclude tuberculosis disease; if the radiograph is normal, the child remains asymptomatic, and Treatment is completed, radiography need not be repeated.

Duration of Therapy for LTBI

For infants, children, and adolescents, the recommended duration of isoniazid therapy is 9 months. Isoniazid is given daily in a single dose. Clinicians who treat LTBI should educate patients and their families about the adverse effects of isoniazid and should prescribe it in monthly allocations, with clinic visits scheduled for monthly face-to-face monitoring. Successful completion of therapy is based on total number of doses taken. When adherence with daily therapy with isoniazid cannot be ensured, twice-a-week DOT can be considered. The twice-weekly regimen should not be prescribed unless each dose is by DOT. Routine determination of serum transaminase values during the 9 months of therapy for LTBI is not indicated.

Therapy for Contacts of Patients With Isoniazid-Resistant M tuberculosis

The incidence of isoniazid resistance among M tuberculosis complex isolates from US patients is approximately 9%. Risk factors for drug resistance are listed in Table 3.85. However, most experts recommend that isoniazid be used to treat LTBI in children unless the child has had contact with a person known to have isoniazid-resistant tuberculosis. If the source case is found to have isoniazid-resistant, rifampin-susceptible organisms, isoniazid should be discontinued and rifampin should be given for a total course of 6 months. A 2-month course of rifampin and pyrazinamide for Treatment of LTBI that once was recommended only for adults no longer is recommended for any age group because of unacceptable hepatotoxicity. Optimal therapy for children with LTBI caused by organisms with resistance to isoniazid and rifampin (ie, MDR) is not known. In these circumstances, multidrug regimens have been used. Drugs to consider include pyrazinamide, a fluoroquinolone, and ethambutol, depending on susceptibility of the isolate. Consultation with a tuberculosis specialist is indicated.

Table 3-85

Treatment of Tuberculosis Disease

The goal of Treatment is to achieve sterilization of the tuberculous lesion in the shortest possible time. Achievement of this goal minimizes the possibility of development of resistant organisms. The major problem limiting successful Treatment is poor adherence to prescribed Treatment regimens. The use of DOT decreases the rates of relapse, Treatment failures, and drug resistance; therefore, DOT is recommended strongly for Treatment of children and adolescents with tuberculosis disease in the United States.

For tuberculosis disease, a 6-month 4-drug regimen consisting of isoniazid, rifampin, pyrazinamide, and ethambutol for the first 2 months and isoniazid and rifampin for the remaining 4 months is recommended for Treatment of pulmonary disease, pulmonary disease with hilar adenopathy, and hilar adenopathy disease in infants, children, and adolescents when an MDR case is not suspected as the source of infection or when drug-susceptibility results are available. Some experts would administer 3 drugs (isoniazid, rifampin, and pyrazinamide) as the initial regimen if a source case has been identified with known pansusceptible M tuberculosis , if the presumed source case has no risk factors for drug-resistant M tuberculosis , or if the source case is unknown but the child resides in an area with low rates of isoniazid resistance. If the chest radiograph shows one or more cavitary lesions and sputum culture remains positive after 2 months of therapy, the duration of therapy should be extended to 9 months. For children with hilar adenopathy in whom drug resistance is not a consideration, a 6-month regimen of only isoniazid and rifampin is considered adequate by some experts.

In the 6-month regimen with 4-drug therapy, isoniazid, rifampin, pyrazinamide, and ethambutol are given once a day for the first 2 weeks by DOT. Between 2 weeks and 2 months of Treatment, these drugs can be given daily or twice or 3 times a week by DOT. After the initial 2-month period, a DOT regimen of isoniazid and rifampin given 2 or 3 times a week is acceptable (see Table 3.82, for doses). Several alternative regimens with differing durations of daily therapy and total therapy have been used successfully in adults and children. These alternative regimens should be prescribed and managed by a specialist in tuberculosis.

When drug resistance is possible (see Table 3.85, above), initial therapy should be adjusted to match the presumed drug susceptibility pattern until drug susceptibility results are available. If an isolate from the pediatric case under Treatment is not available, drug susceptibilities can be inferred by the drug susceptibility pattern of isolates from the adult source case. Data for guiding drug selection may not be available for foreign-born children or in circumstances of international travel. If this information is not available, a 4-drug initial regimen is recommended with close monitoring for clinical response.

Therapy for Drug-Resistant Tuberculosis Disease

Drug resistance is most common in the following: (1) people born in areas such as Russia and the former Soviet Union, Asia, Africa, and Latin America; (2) people previously treated for tuberculosis disease; and (3) contacts, especially children, with tuberculosis disease whose source case is a person from one of these groups (see also Table 3.85). Most cases of pulmonary tuberculosis in children that are caused by an isoniazid-resistant but rifampin- and pyrazinamide-susceptible strain of M tuberculosis complex can be treated with a 6-month regimen of rifampin, pyrazinamide, and ethambutol. For cases of MDR tuberculosis disease, the Treatment regimen should include at least 4 antituberculosis drugs to which the organism is susceptible. In cases of tuberculosis with isoniazid- and rifampin-resistant strains, 6-month drug regimens are not recommended. A regimen lasting 12 to 24 months of therapy usually is necessary for cure. Regimens in which drugs are administered 2 or 3 times per week also are not recommended for drug-resistant disease; daily DOT is critical to cure children with drug-resistant tuberculosis disease and to prevent emergence of further resistance.

Extrapulmonary M tuberculosis Tuberculosis Disease

In general, extrapulmonary tuberculosis-with the exception of meningitis-can be treated with the same regimens as used for pulmonary tuberculosis. For suspected drug-susceptible tuberculous meningitis, daily Treatment with isoniazid, rifampin, pyrazinamide, and ethambutol or ethionamide, if possible, or an aminoglycoside should be initiated. When susceptibility to all drugs is established, the ethambutol, ethionamide, or aminoglycoside can be discontinued. Pyrazinamide is given for a total of 2 months and isoniazid and rifampin are given for a total of 9 to 12 months. Isoniazid and rifampin can be given daily or 2 or 3 times per week after the first 2 months of Treatment.

Corticosteroids

The evidence supporting adjuvant Treatment with corticosteroids for children with tuberculosis disease is incomplete. Corticosteroids are indicated for children with tuberculous meningitis, because corticosteroids decrease rates of mortality and long-term neurologic impairment. Corticosteroids can be considered for children with pleural and pericardial effusions (to hasten reabsorption of fluid), severe miliary disease (to mitigate alveolocapillary block), endobronchial disease (to relieve obstruction and atelectasis), and abdominal tuberculosis (to decrease the risk of strictures). Corticosteroids should be given only when accompanied by appropriate antituberculosis therapy. Most experts consider 2 mg/kg per day of prednisone (maximum, 60 mg/day) or its equivalent for 4 to 6 weeks followed by tapering to be appropriate.

Tuberculosis Disease and HIV Infection

Adults and children with HIV infection have an increased incidence of tuberculosis disease. Hence, HIV testing is indicated for all patients with tuberculosis disease . The Clinical Manifestations and radiographic appearance of tuberculosis disease in children with HIV infection tend to be similar to those in immunocompetent children, but manifestations in these children can be more severe and unusual and can include extrapulmonary involvement of multiple organs. In HIV-infected patients, a TST result of 5-mm induration or more is considered positive (see Table 3.79); however, a negative TST result attributable to HIV-related immunosuppression also can occur. Specimens for culture should be obtained from all HIV-infected children with suspected tuberculosis.

Most HIV-infected adults with drug-susceptible tuberculosis respond well to antituberculosis drugs when appropriate therapy is given early. However, optimal therapy for tuberculosis in children with HIV infection has not been established. Treating tuberculosis in an HIV-infected child is complicated by antiretroviral drug interactions with the rifamycins and overlapping toxicities caused by antiretroviral drugs and medications used to treat tuberculosis. Therapy always should include at least 4 drugs initially and be continued for at least 9 months. Isoniazid, rifampin, and pyrazinamide, usually with ethambutol or an aminoglycoside, should be given for at least the first 2 months. A 3-drug regimen can be used once drug-resistant tuberculosis disease is excluded. Rifampin may be contraindicated in people who are receiving highly active antiretroviral therapy. Rifabutin can be substituted for rifampin in some circumstances. Consultation with a specialist who has experience in managing HIV-infected patients with tuberculosis is advised strongly.

Evaluation and Monitoring of Therapy in Children and Adolescents

Careful monthly monitoring of the clinical and bacteriologic responses to therapy is important. With DOT, clinical evaluation is an integral component of each visit for drug administration. For patients with pulmonary tuberculosis, chest radiographs should be obtained after 2 months of therapy to evaluate response. Even with successful 6-month regimens, hilar adenopathy can persist for 2 to 3 years; normal radiographic findings are not necessary to discontinue therapy. Follow-up chest radiography beyond termination of successful therapy usually is not necessary unless clinical deterioration occurs.

If therapy has been interrupted, the date of completion should be extended. Although guidelines cannot be provided for every situation, factors to consider when establishing the date of completion include the following: (1) length of interruption of therapy; (2) time during therapy (early or late) when interruption occurred; and (3) the patient's clinical, radiographic, and bacteriologic status before, during, and after interruption of therapy. The total doses administered by DOT should be calculated to guide the duration of therapy. Consultation with a specialist in tuberculosis is advised.

Untoward effects of isoniazid therapy, including severe hepatitis in otherwise healthy infants, children, and adolescents, are rare. Routine determination of serum transaminase concentrations is not recommended. However, for children with severe tuberculosis disease, especially children with meningitis or disseminated disease, transaminase concentrations should be monitored approximately monthly during the first several months of Treatment. Other indications for testing include the following: (1) having concurrent or recent liver or biliary disease; (2) being pregnant or in the first 6 weeks postpartum; (3) having clinical evidence of hepatotoxic effects; or (4) concurrently using other hepatotoxic drugs (eg, anticonvulsant or HIV agents). In most other circumstances, monthly clinical evaluations to observe for signs or symptoms of hepatitis and other adverse effects of drug therapy without routine monitoring of transaminase concentrations is appropriate follow-up. In all cases, regular physician-patient contact to assess drug adherence, efficacy, and adverse effects is an important aspect of management. Patients should be advised to call a physician immediately if signs of adverse effects, in particular hepatotoxicity (eg, vomiting, abdominal pain, jaundice), develop.

Immunizations

Patients who are receiving Treatment for tuberculosis can be given measles and other age-appropriate attenuated live-virus vaccines unless they are receiving high-dose corticosteroids, are severely ill, or have other specific contraindications to immunization.

Tuberculosis During Pregnancy and Breastfeeding

Tuberculosis Treatment during pregnancy varies because of the complexity of management decisions. During pregnancy, if tuberculosis disease is diagnosed, a regimen of isoniazid, rifampin, and ethambutol is recommended. Pyrazinamide commonly is used in a 3- or 4-drug regimen, but safety during pregnancy has not been established. At least 6 months of therapy is indicated for drug-susceptible tuberculosis disease if pyrazinamide is used; at least 9 months of therapy is indicated if pyrazinamide is not used. Prompt initiation of therapy is mandatory to protect mother and fetus.

Asymptomatic pregnant women with a positive TST or IGRA result, normal chest radiographic findings, and recent contact with a contagious person should be considered for isoniazid therapy. The recommended duration of therapy is 9 months. Therapy in these circumstances should begin after the first trimester. Pyridoxine supplementation is indicated for all pregnant and breastfeeding women receiving isoniazid.

Isoniazid, ethambutol, and rifampin are relatively safe for the fetus. The benefit of ethambutol and rifampin for therapy of tuberculosis disease in the mother outweighs the risk to the infant. Because streptomycin can cause ototoxic effects in the fetus, it should not be used unless administration is essential for effective Treatment. The effects of other second-line drugs on the fetus are unknown.

Although isoniazid is secreted in human milk, no adverse effects of isoniazid on nursing infants have been demonstrated (see Human Milk). Breastfed infants do not require pyridoxine supplementation unless they are receiving isoniazid.

Congenital Tuberculosis

Women who have only pulmonary tuberculosis are not likely to infect the fetus but can infect their infant after delivery. Congenital tuberculosis is rare, but in utero infections can occur after maternal bacillemia.

If a newborn infant is suspected of having congenital tuberculosis, a TST, chest radiography, lumbar puncture, and appropriate cultures should be performed promptly. The TST result usually is negative in newborn infants with congenital or perinatally acquired infection. Only case reports of results of IGRAs in selected newborn infants have been published, and IGRAs should not be substituted for the TST in newborn infants. Hence, regardless of the TST results, Treatment of the infant should be initiated promptly with isoniazid, rifampin, pyrazinamide, and an aminoglycoside (eg, amikacin). The placenta should be examined histologically for granulomata and AFB, and a specimen should be cultured for M tuberculosis complex. The mother should be evaluated for presence of pulmonary or extrapulmonary disease, including uterine tuberculosis disease. If the maternal physical examination and chest radiographic findings support the diagnosis of tuberculosis disease, the newborn infant should be treated with regimens recommended for tuberculosis disease. If meningitis is confirmed, corticosteroids should be added (see Corticosteroids). Drug susceptibility testing of the organism recovered from the mother or household contact, infant, or both should be performed.

Management of the Newborn Infant Whose Mother (or Other Household Contact) Has LTBI or Tuberculosis Disease

Management of the newborn infant is based on categorization of the maternal (or household contact) infection. Although protection of the infant from exposure and infection is of paramount importance, contact between infant and mother should be allowed when possible. Differing circumstances and resulting recommendations are as follows:

• Mother (or household contact) has a positive TST or IGRA result and normal chest radiographic findings . If the mother (or household contact) is asymptomatic, no separation is required. The mother usually is a candidate for Treatment of LTBI after the initial postpartum period. The newborn infant needs no special evaluation or therapy. Because the positive TST or IGRA result could be a marker of an unrecognized case of contagious tuberculosis within the household, other household members should have a TST or IGRA and further evaluation, but this should not delay the infant's discharge from the hospital. These mothers can breastfeed their infants.


• Mother (or household contact) has clinical signs and symptoms or abnormal findings on chest radiograph consistent with tuberculosis disease . Cases of suspected or proven tuberculosis disease in mothers (or household contacts) should be reported immediately to the local health department, and investigation of all household members should start within 7 days. If the mother has tuberculosis disease, the infant should be evaluated for congenital tuberculosis (see Congenital Tuberculosis), and the mother should be tested for HIV infection. The mother (or household contact) and the infant should be separated until the mother (or household contact) has been evaluated and, if tuberculosis disease is suspected, until the mother (or household contact) and infant are receiving appropriate antituberculosis therapy, the mother wears a mask, and the mother understands and is willing to adhere to infection-Control Measures. Once the infant is receiving isoniazid, separation is not necessary unless the mother (or household contact) has possible MDR tuberculosis disease or has poor adherence to Treatment and DOT is not possible. In this circumstance, the infant should be separated from the mother (or household contact), and BCG immunization should be considered for the infant. If the mother is suspected of having MDR tuberculosis disease, an expert in tuberculosis disease Treatment should be consulted. Women with tuberculosis disease who have been treated appropriately for 2 or more weeks and who are not considered contagious can breastfeed.

If congenital tuberculosis is excluded, isoniazid is given until the infant is 3 or 4 months of age, when a TST should be performed. If the TST result is positive, the infant should be reassessed for tuberculosis disease. If tuberculosis disease is excluded, isoniazid should be continued for a total of 9 months. The infant should be evaluated at monthly intervals during Treatment. If the TST result is negative at 3 to 4 months of age and the mother (or household contact) has good adherence and response to Treatment and no longer is contagious, isoniazid is discontinued.

• Mother (or household contact) has abnormal findings on chest radiography but no evidence of tuberculosis disease . If the chest radiograph of the mother (or household contact) appears abnormal but is not characteristic of tuberculosis disease and the history, physical examination, and sputum smear indicate no evidence of tuberculosis disease, the infant can be assumed to be at low risk of tuberculosis infection and need not be separated from the mother (or household contact). The mother and her infant should receive follow-up care and the mother should be treated for LTBI. Other household members should have a TST or IGRA and further evaluation.

Most children with tuberculosis disease, especially children younger than 10 years, are not contagious. Exceptions are the following: (1) children with cavitary pulmonary tuberculosis; (2) children with positive sputum AFB smears; (3) children with laryngeal involvement; (4) children with extensive pulmonary infection; or (5) children with congenital tuberculosis undergoing procedures that involve the oropharyngeal airway (eg, endotracheal intubation). In these instances, isolation for tuberculosis or AFB are indicated until effective therapy has been initiated, sputum smears demonstrate a diminishing number of organisms, and cough is abating. Children with no cough and negative sputum AFB smears can be hospitalized in an open ward. Infection-Control Measures for hospital personnel exposed to contagious patients should include the use of personally "fitted" and "sealed" particulate respirators for all patient contacts (see Infection Control for Hospitalized Children). The contagious patient should be placed in an airborne infection isolation room in the hospital.

The major concern in infection control relates to adult household members and contacts who can be the source of infection. Visitation should be limited to people who have been evaluated medically. Household members and contacts should be managed with tuberculosis precautions when visiting until they are demonstrated not to have contagious tuberculosis. Nonadherent household contacts should be excluded from hospital visitation until evaluation is complete and tuberculosis disease is excluded or Treatment has rendered source cases noncontagious.

Infections with M bovis account for approximately 1% to 2% of tuberculosis cases in the United States. Children who come from countries where M bovis is prevalent in cattle or whose parents come from those countries are more likely to be infected. Most infections in humans are transmitted from cattle by unpasteurized milk and its products, such as fresh cheese, although human-to-human transmission by the airborne route has been documented. In children, M bovis more commonly causes cervical lymphadenitis, intestinal tuberculosis disease, and meningitis. In adults, latent M bovis infection can progress to advanced pulmonary disease, with a risk of transmission to others.

The diagnosis of tuberculosis caused by M bovis infection requires a culture isolate. The commonly used methods for identifying M tuberculosis complex do not distinguish M bovis from M tuberculosis, M africanum , and BCG; M bovis is identified in clinical laboratories routinely by its resistance to pyrazinamide. However, this approach can be unreliable, and species confirmation at a reference laboratory should be requested when M bovis is suspected. Resistance to first-line drugs in addition to pyrazinamide has been reported. BCG rarely is isolated from pediatric clinical specimens; however, it should be suspected from the characteristic lesions or localized BCG suppuration or draining lymphadenitis in children who have received BCG vaccine. Only a reference laboratory can distinguish an isolate of BCG from an isolate of M bovis .

Therapy for M bovis Disease

Controlled clinical trials for Treatment of M bovis disease have not been conducted, and Treatment recommendations for M bovis disease in adults and children are based on results from Treatment trials for M tuberculosis disease. Although most strains of M bovis are pyrazinamide-resistant and resistance to other first-line drugs has been reported, MDR strains are rare. Initial therapy should include 3 or 4 drugs besides pyrazinamide that would be used to treat disease from M tuberculosis infection. For isoniazid- and rifampin-susceptible strains, a total Treatment course of at least 9 to 12 months is recommended.

Parents should be counseled about the many infectious diseases transmitted by unpasteurized milk and its products, and parents who might import traditional dairy products from countries where M bovis infection is prevalent in cattle should be advised against giving those products to their children. When people are exposed to an adult who has pulmonary disease caused by M bovis infection, they should be evaluated by the same methods as other contacts to contagious tuberculosis.

Control of tuberculosis disease in the United States requires collaboration between health care professionals and health department personnel, obtaining a thorough history of exposure(s) to people with infectious tuberculosis, timely and effective contact investigations, proper interpretation of TST or IGRA results, and appropriate antituberculosis therapy, including DOT services. A plan to control and prevent extensively drug-resistant tuberculosis has been published.² Eliminating ingestion of unpasteurized dairy products will prevent most M bovis infection.

Management of Contacts, Including Epidemiologic Investigation

Children with a positive TST or IGRA result or tuberculosis disease should be the starting point for epidemiologic investigation by the local health department. Close contacts of a TST- or IGRA-positive child should have a TST or IGRA, and people with a positive TST or IGRA result or symptoms consistent with tuberculosis disease should be investigated further. Because children with tuberculosis usually are not contagious unless they have an adult-type multibacillary form of pulmonary or laryngeal disease, their contacts are not likely to be infected unless they also have been in contact with the same adult source case. After the presumptive adult source of the child's tuberculosis is identified, other contacts of that adult should be evaluated.

Therapy for Contacts

Children and adolescents exposed to a contagious case of tuberculosis disease should have a TST or IGRA and an evaluation for tuberculosis disease (chest radiography and physical examination). For exposed contacts with impaired immunity (eg, HIV infection) and all contacts younger than 4 years of age, isoniazid therapy should be initiated, even if the TST result is negative, once tuberculosis disease is excluded (see Therapy for LTBI). Infected people can have a negative TST or IGRA result because a cellular immune response has not yet developed or because of cutaneous anergy. People with a negative TST or IGRA result should be retested 8 to 10 weeks after the last exposure to a source of infection. If the TST or IGRA result still is negative in an immunocompetent person, isoniazid is discontinued. If the contact is immunocompromised and LTBI cannot be excluded, Treatment should be continued for 9 months. If a TST or IGRA result of a contact becomes positive, isoniazid should be continued for 9 months.

Child Care and Schools

Children with tuberculosis disease can attend school or child care if they are receiving therapy (see Children in Out-of-Home Child Care). They can return to regular activities as soon as effective therapy has been instituted, adherence to therapy has been documented, and clinical symptoms have diminished. Children with LTBI can participate in all activities whether they are receiving Treatment or not.

BCG Vaccines

The bacille Calmette-Guérin (BCG) vaccine is a live vaccine originally prepared from attenuated strains of M bovis . Use of BCG vaccine is recommended by the Expanded Programme on Immunization of the World Health Organization for administration at birth (see Table 1.3) and is used in more than 100 countries. BCG vaccine is used to reduce the incidence of disseminated and other life-threatening manifestations of tuberculosis in infants and young children. Although BCG immunization appears to decrease the risk of serious complications of tuberculosis disease in children, the various BCG vaccines used throughout the world differ in composition and efficacy.

Two meta-analyses of published clinical trials and case-control studies concerning the efficacy of BCG vaccines concluded that BCG vaccine has relatively high protective efficacy (approximately 80%) against meningeal and miliary tuberculosis in children. The protective efficacy against pulmonary tuberculosis differed significantly among the studies, precluding a specific conclusion. Protection afforded by BCG vaccine in one meta-analysis was estimated to be 50%. Two BCG vaccines, one manufactured by Organon Teknika Corporation and the other by sanofi pasteur, are licensed in the United States. Comparative evaluations of these and other BCG vaccines have not been performed.

Indications

In the United States, administration of BCG vaccine should be considered only in limited and select circumstances, such as unavoidable risk of exposure to tuberculosis and failure or unfeasibility of other control methods. Recommendations for use of BCG vaccine for control of tuberculosis among children and health care professionals have been published by the Advisory Committee on Immunization Practices of the CDC and the Advisory Council for the Elimination of Tuberculosis.¹ For infants and children, BCG immunization should be considered only for people with a negative TST result who are not infected with HIV in the following circumstances:

• The child is exposed continually to a person or people with contagious pulmonary tuberculosis resistant to isoniazid and rifampin, and the child cannot be removed from this exposure.


• The child is exposed continually to a person or people with untreated or ineffectively treated contagious pulmonary tuberculosis, and the child cannot be removed from such exposure or given antituberculosis therapy.

Careful assessment of the potential risks and benefits of BCG vaccine and consultation with personnel in local tuberculosis control programs are recommended strongly before use of BCG vaccine.
Healthy infants from birth to 2 months of age may be given BCG vaccine without a TST unless congenital infection is suspected; thereafter, BCG vaccine should be given only to children with a negative TST result.

Adverse Reactions

Uncommonly (1%-2% of immunizations), BCG vaccine can result in local adverse reactions, such as subcutaneous abscess and regional lymphadenopathy, which generally are not serious. One rare complication, osteitis affecting the epiphysis of long bones, can occur as long as several years after BCG immunization. Disseminated fatal infection occurs rarely (approximately 2 per 1 million people), primarily in people who are immunocompromised severely. Antituberculosis therapy is recommended to treat osteitis and disseminated disease caused by BCG vaccine. Pyrazinamide is not believed to be effective against BCG and should not be included in Treatment regimens. Most experts do not recommend Treatment of draining skin lesions or chronic suppurative lymphadenitis caused by BCG vaccine, because spontaneous resolution occurs in most cases. Large-needle aspiration of suppurative lymph nodes can hasten resolution. People with complications caused by BCG vaccine should be referred for management, if possible, to a tuberculosis expert.

Contraindications

People with burns, skin infections, and primary or secondary immunodeficiencies, including HIV infection, should not receive BCG vaccine. Because an increasing number of cases of localized and disseminated BCG have been described in infants and children with HIV infection, the World Health Organization no longer recommends BCG in healthy, HIV-infected children. Use of BCG vaccine is contraindicated for people receiving immunosuppressive medications, including high-dose corticosteroids (see Corticosteroids). Although no untoward effects of BCG vaccine on the fetus have been observed, immunization during pregnancy is not recommended.

Reporting of Cases

Reporting of suspected and confirmed cases of tuberculosis disease is mandated by law in all states. A diagnosis of LTBI or tuberculosis disease in a child is a sentinel event representing recent transmission of M tuberculosis in the community. Physicians should assist local health department personnel in the search for a source case and others infected by the source case. Members of the household, such as relatives, babysitters, au pairs, boarders, domestic workers, and frequent visitors or other adults, such as child care providers and teachers with whom the child has frequent contact, potentially are source cases.

a . These drugs should be used in consultation with a specialist in tuberculosis.

b . Dose adjustment in renal insufficiency.

c . Levofloxacin is not approved for use in children younger than 18 years of age; its use in younger children necessitates assessment of the potential risks and benefits (see Antimicrobial Agents and Related Therapy).

1 . Centers for Disease Control and Prevention. The role of BCG vaccine in the prevention and control of tuberculosis in the United States: a joint statement by the Advisory Committee for the Elimination of Tuberculosis and the Advisory Committee on Immunization Practices. MMWR Recomm Rep . 1996;45(RR-4):1-18

2 . Centers for Disease Control and Prevention. Plan to combat extensively drug-resistant tuberculosis: recomendations of the Federal Tuberculosis Task Froce. MMWR Recomm Rep . 2009;58(RR-3):1-43

3 . National Tuberculosis Controllers Association and Centers for Disease Control and Prevention. Guidelines for the investigation of contacts of persons with infectious tuberculosis. Recommendations from the National Tuberculosis Controllers Association and CDC. MMWR Recomm Rep . 2005;54(RR-15):1-47