Tuberculosis (TB) remains one of the most devastating infectious diseases in the world. The World Health Organization (WHO) estimates that approximately one-third of the world's population is infected with Mycobacterium tuberculosis, with 88 million new cases from 1990 to 1999. Thirty million people worldwide died of TB during this period, including 2.9 million persons coinfected with human immunodeficiency virus and acquired immunodeficiency syndrome (HIV/AIDS). Between 1990 and 1999, the incidence worldwide increased from 143 cases per 100,000 persons to 163 cases per 100,000 persons.1 Factors contributing to the global increase in TB included population growth and the emerging HIV epidemic. The WHO and the international public health community have prioritized the control, prevention, and treatment of TB during this millennium.
During a resurgence in the United States that peaked in 1992,2 the incidence of TB reached 10.4 per 100,000 persons. Case rates were highest in ethnic minorities, with highs of 46.6 and 31.7 per 100,000 persons for Asians and non-Hispanic blacks, respectively. Non-Hispanic whites had the lowest case rate of 4.0 per 100,000 persons. Several factors contributed to this resurgence, including a large influx of foreign-born persons during the late 1970s, the HIV epidemic, homelessness, crowded living conditions, and the dismantling of the public health structure for TB care. In addition, there was an increase in multidrug-resistant TB, denned as resistance to isoniazid and rifampin. This included both primary acquisition of multidrug-resistant strains attributed to health care-associated infection and overcrowding in homeless shelters and secondary cases due to nonadherence to TB therapy. Multidrug-resistant TB is also associated with higher mortality rates and costs. Since 1992, case rates, including cases of multidrug-resistant TB, have steadily fallen in the United States because of a huge influx of financial and technical support to health departments and hospitals. The case rate of TB dropped by 38% from 1992 to 1997.2 Foreign birth remains the major risk factor for TB in the United States.2
During the 1990s, the case rates of pediatrie TB, defined as TB occurring among children younger than 15 years, also increased, although the total number of cases was a fraction of those in adults. TB in adults may be primary or reactivation disease, whereas TB in children is a sentinel event indicating recent transmission of M. tuberculosis from an infected adult or adolescent source. Thus, risk factors for TB in children reflect contact with adults who are at risk for TB (Table 1).
Adults at Risk for Tuberculosis
Factors That Increase the Risk of Progression to Active Tuberculosis In Children
RISK FACTORS FOR INFECTION AND PROGRESSION TO ACTIVE DISEASE
Infection with M. tuberculosis and progression of infection to active TB are dependent on many factors. The most important factor leading to infection is close contact with an infected adult or adolescent source case. Individuals whose sputa are smear positive for acid-fast bacilli are more infectious than those whose sputa are only culture positive.3 Cavitary lesions and forceful coughing increase the risk of dissemination, as do close contact and closed environments.4 When there is coinfection with HIV and AIDS, individuals may be contagious even with normal findings on chest radiograph or the absence of cavitary lesions.5
Once a child is infected with TB, disease progression is determined by several factors (Table 2). The risk of progression from latent TB infection to active TB is inversely proportional to age; 43% of children younger than 1 year, 24% of children 1 to 5 years old, and 5% to 15% of adolescents older than 15 years have TB without treatment for latent TB infection.6 Thus, the risk is greatest for children younger than 1 year and such children also have the greatest risk for progression to extrapulmonary disease (eg, meningeal and miliary TB). The risk for active TB is highest in all patients within the first year after initial TB infection. Children who are malnourished and immunocompromised are also at higher risk for progression to active TB.
TB describes a wide range of clinical illness caused by the M. tuberculosis complex, which consists of M. tuberculosis and M. bovis. M. tuberculosis are nonmotile, aerobic, non-spore-forming bacilli with a cell wall that has a high content of high-molecular-weight lipids. Mycobacteria species stain poorly with gram stain, but the cell wall avidly binds to carbolfuchsin dye and resists decolorization with acid-alcohol during Ziehl-Neelsen staining, hence the term acid-fast bacilli.
Laboratory evaluation of specimens for M. tuberculosis includes staining and microscopic evaluation for acid-fast bacilli, growth of mycobacteria, species identification, and susceptibility testing. The limit of detection of acid-fast bacilli staining is generally IO3 to IO4 acid-fast bacilli per milliliter of specimen. Thus, gastric aspirates are frequently negative for acid-fast bacilli because of the low concentration of organisms. Specimens obtained by bronchoalveolar lavage are no better than properly obtained gastric aspirates at detecting M. tuberculosis.7 Staining sputa for acid-fast bacilli is a low-cost, rapid screening test and is useful for identifying potentially infectious patients, but a negative result does not rule out TB.
Culturing specimens for M. tuberculosis remains the gold standard for diagnosing TB. Unfortunately, mycobacteria species grow slowly even on specialized solid media (eg, LowensteinJensen and Middlebrook medium) and require 3 to 6 weeks for visible colonies to appear. A radiometric culture system is preferable because growth of mycobacteria species can be detected by measuring radioactive 14CO2 metabolized from 14C-labeled palmitic acid in the liquid medium within 7 to 14 days.8 Gastric aspirates, the most commonly processed samples from children, are positive for M. tuberculosis in only 20% to 30% of cases, because of the low organism burden and killing of M. tuberculosis by low gastric pH.
Because of the slow growth of M. tuberculosis, more rapid methods to identify this pathogen have been developed. Once growth is detected, nucleicacid probe hybridization assays can target the ribosomal RNA and identify M. tuberculosis, M. avium, M. kansasii, and M. gordonae. This assay allows identification from clinical specimens and solid and liquid culture media.9·10 Finally, nucleic-acid amplification of genomic DNA by polymerase chain reaction (PCR) can directly detect M. tuberculosis in clinical specimens with the advantage of not requiring growth of mycobacteria. This detects M. tuberculosis in extrapulmonary specimens with sensitivities ranging from 38% to 72%" and in acid-fast bacilli smear-negative sputum specimens with sensitivities ranging from 50% to 80%.12·13 However, commercially available PCR has little utility in detecting M. tuberculosis in gastric aspirates, again related to a low organism burden.14 Specimens that are positive by these more rapid methods still require culture for susceptibility testing.
Susceptibility to antimycobacterial drugs is performed in two stages. First-line drugs, including isoniazid, rifampin, pyrazinamide, and ethambutol, are studied initially. If an organism is susceptible to all of these agents, then no further testing is required. If an organism is found to be mulridrug-resistant TB, testing of second-line agents is performed. Again, because of the slow growth of M. tuberculosis, such testing requires at least 2 weeks for first-line agents and 2 to 4 weeks for second-line agents.
The lungs are usually the route of infection with M. tuberculosis following contact with an infectious source. Following inhalation of aerosolized droplets laden with M. tuberculosis, the bacilli infect alveoli in well-aerated portions of the lungs and cause inflammation. An aggregation of inflammatory cells and reactive fibrosis in the lungs results in the Ghon focus, which may become calcified (latent TB infection). Although most bacilli are contained within the Ghon focus, some escape containment and travel via lymphatics to the regional lymph nodes, causing further inflammation as macrophages and lymphocytes are recruited to the lymph nodes. The combination of the Ghon focus and enlarged hilar and paratracheal lymph nodes is known as the primary complex (TB). During primary infection, mycobacteria may seed other sites, including the meninges, the brain, or the gastrointestinal tract, via the lymphohematogenous route (extrapulmonary TB). The bacilli may also spread via the endobronchial route to the adjacent areas of the lungs. After the primary infection, foci of infection may undergo fibrosis and contain only a few dormant bacilli. These organisms have the potential to reactivate later or with immunosuppression.
CLINICAL PRESENTATIONS AND MANAGEMENT OF M. TUBERCULOSIS
Management and treatment of latent TB infection and active TB depend on the patient's stage in the clinical evolution of the disease. The WHO has classified the clinical stages of TB infection and disease to standardize epidemiology and treatment (Table 3).15·16
The Centers for Disease Control and Prevention case definitions of TB are based on laboratory isolation of M. tuberculosis from clinical specimens (laboratory-confirmed case) or clinical signs and symptoms with supporting radiographie studies (clinical case). Thus, a laboratory-confirmed diagnosis of TB requires isolation and identification of M. tuberculosis by acidfast bacilli staining, culture, or DNA or PCR probe from clinical specimens. Clinical case definitions often are used to diagnose TB, especially in children, because a laboratory diagnosis may not be readily available and may take weeks or cultures may be negative. The criteria for the clinical case definition of TB include (1) a positive result on the tuberculin skin test (TST); (2) clinical signs and symptoms consistent with M. tuberculosis; (3) abnormal, unstable findings on chest radiograph compatible with TB; (4) a full diagnostic evaluation; and (5) resolution of signs and symptoms of TB with two or more mycobacterial drugs.17
Modified International Classification of Tuberculosis
Although neither the laboratory-confirmed nor the clinical case definition considers the epidemiology of TB, it is critical to assess children with suspected TB for risk factors. Such risk factors include contact with an adult or an adolescent with TB, foreign birth, foreign travel, and ingestion of unpasteurized milk. All patients and their families should be interviewed for these risk factors.
The most common clinical presentation of TB in children is pulmonary TB because more than 75% of children have involvement of the lungs. 17 Approximately 50% of children with active pulmonary disease are asymptomatic, especially when diagnosed early such as occurs during a contact investigation of an infectious source case. However, younger children are more likely to be symptomatic.18'19 The remainder of children with pulmonary TB have nonspecific symptoms and signs, including cough, fever, weight loss or failure to thrive, malaise, loss of energy, vomiting, diarrhea, and, rarely, night sweats. Enlarged lymph nodes compressing the esophagus and local nerves may cause odynophagia and paralysis of the vocal cords, the phrenic nerve, or both.17'18·20 On physical examination, children with TB can exhibit bronchial breathing, crackles, diminished breath sounds, strider, and wheezing.
A TST is generally useful, particularly if results are positive. Currently, only TSTs performed by the Mantoux test are acceptable. The Tine test is nonspecific, not standardized, and inconvenient because a positive result must be followed by the Mantoux TST, which adds cost. Results of a TST are usually positive in children with active TB, but the size and interpretation of the TST result may vary. For instance, a TST result of 5 mm or greater induration in a child with a suggestive chest radiograph is considered positive.21 In approximately 10% of children with active TB, the TST result may be negative because of young age, overwhelming TB, or immunosuppression.
Anteroposterior and lateral chest radiographs are critical to diagnose pulmonary TB. The most common radiographie finding in children is mediasrinal lymphadenopathy. A lateral view may be more useful to detect adenopathy. Additional features include airway compresion (eg, consolidation, atelectasis, air trapping, or bronchopneumonia), pleural effusion, a miliary pattern, and cavitation in adolescents. If the chest radiograph is equivocal, a computed tomography scan of me chest may define adenopathy, the thymus, or other anatomic abnormalities more dearly.14
Most young children do not produce sputum and rarely have cavitary disease. Thus, alternative respiratory tract specimens must be obtained. In young children, respiratory tract secretions may be collected by sputum induction,22 bronchoscopy, or early morning gastric aspirates with lavage. However, if a source case is known, treatment can be guided by the antimy cob acte rial susceptibilities of the source case's M. tuberculosis isolate. Induced sputum and bronchoscopy specimens must be obtained in negative pressure rooms with health care workers wearing appropriate masks. All specimens for M. tuberculosis must be delivered to the laboratory immediately for processing, because the low gastric pH may kill viable mycobacteria.
The response to treatment is variable, Although most treated patients improve with resolution of their symptoms, the conditions of some initially may worsen due to an inflammatory response related to treatment. Such symptoms may include fever and worsening respiratory symptoms related to lymph node enlargement. Chest radiographs also are useful while monitoring patients receiving treatment for TB. Studies are obtained at diagnosis, 1 to 2 months after therapy has begun, and then every 2 to 3 months until the completion of therapy. Adenopathy may not normalize until months after therapy is completed. Thus, some physicians obtain studies 6 to 12 months later in such children whose initial cultures were negative, who did not have a known source case, or who were treated for multidrug-resistant TB to ensure radiologie resolution.
Extrapulmonary TB occurs in fewer than 10% of children with TB.17 The most common presentations are peripheral lymphadenopathy, tuberculous meningitis, miliary TB, and bone and joint disease.
Peripheral Lymphadenopathy. Peripheral lymphadenopathy is the most common manifestation of extrapulmonary TB. Lymphadenopathy commonly involves the anterior and the posterior cervical and supraclavicular lymph nodes and is secondary to either a primary source (eg, the lungs) with spread to the mediastinal and supraclavicular lymph nodes or hematogenous spread to distant lymph nodes. TB lymphadenitis presents as an enlarged, nontender, firm lymph node with minimal fluctuance or erythema. If a group of lymph nodes is involved, they are usually matted together. There are no or few constitutional symptoms. Scrofula is a well-described form of TB lymphadenitis; infected superficial and deep cervical lymph nodes present as nodules, plaques, ulcers, and sinuses. Fine needle aspiration or biopsy of the lymph nodes is needed to confirm the diagnosis by culture because cervical lymphadenopathy is more commonly caused by nontuberculous mycobacteria, which cannot be distinguished from M. tuberculosis by acid-fast bacilli staining or histopathology. The TB gene probe or PCR may be performed on biopsy specimens from lymph nodes.
Tuberculous Meningitis and Tuberculoma. Tuberculous meningitis causes significant longterm morbidity and mortality in children. It occurs more commonly in children younger than 6 years. In children, seeding of the brain and meninges occurs during primary infection or during miliary dissemination. The clinical presentation is usually insidious and a high index of suspicion is required to make the diagnosis. Symptoms may include fever, vomiting, behavioral and personality changes, headache, meningismus, and seizures. Physical signs may include nuchal rigidity, cranial nerve palsies (particularly IV and VI), altered deep tendon reflexes, and focal neurologic signs. Extensive central nervous system damage may result in extrapyramidal signs, coma, and death. Among children with meningeal TB, 40% to 50% have other foci of TB, most commonly miliary disease.23
Evaluation of the cerebrospinal fluid is critical to diagnose tuberculous meningitis. The opening pressure is usually elevated, the cell count is elevated with a lymphocytic predominance, the protein level is significantly elevated, and the glucose level is characteristically low. Early on, the cerebrospinal fluid may be normal and a repeat evaluation is warranted if the suspicion for meningitis is high. Although the identification of M. tuberculosis in the cerebrospinal fluid remains the gold standard for diagnosing tuberculous meningitis, the yield is low, particularly if less than 10 mL of cerebrospinal fluid is evaluated. However, cultures from other sites may be diagnostic and PCR can also be performed on cerebrospinal fluid. A computed tomography scan or a magnetic resonance imaging scan with and without contrast of the brain may demonstrate brain edema, hydrocephalus, ventriculomegaly, basal meningeal enhancement, periventricular lucencies, or a tuberculoma.
Miliary Tuberculosis. Miliary TB occurs in a small percentage of children with active TB. Similar to tuberculous meningitis, miliary TB most often occurs during primary infection. Symptoms in children are usually subacute and may include fever, night sweats, malaise, weight loss, and rigors. With dissemination, headache, seizures, or changes in personality suggest meningitis; abdominal pain suggests peritonitis; and respiratory distress and chest pain suggest pleurisy or diffuse lung disease. Physical signs are usually nonspecific, although generalized lymphadenopathy and hepatomegaly may be found. Cutaneous eruptions and choroidal tubercles of the retina also are present. A normal peripheral white blood cell count and anemia are nonspecific laboratory findings, and hyponatremia from inappropriate secretion of antidiuretic hormone or Addison's disease may be present. Liver enzymes and alkaline phosphatase may be elevated. The typical appearance of countless small nodules ("millet seeds") on chest radiographs in this clinical context strongly suggests miliary TB.
Musculoskeletal Tuberculosis. Bone biopsy and culture are critical to ensure appropriate diagnosis and treatment of musculoskeletal TB. Such patients should be seen by a pediatrie orthopedist. Serial drainage of synovial fluid may be needed for management of TB arthritis. Spinal stabilization, surgical management, or both is critical with Potf s disease (spinal TB).
Investigations Conducted by Health Departments to Diagnose Tuberculosis
TREATMENT OF TUBERCULOSIS
The treatment of children with active TB and latent TB infection includes both medical treatment of the individual patient and the activities of the health department. Initially, empiric therapy is provided and, if the child has a positive result on culture for M. tuberculosis or the susceptibility of the source case is known, more specific therapy can be administered. The site of infection (eg, pulmonary vs extrapulmonary) and the presence or absence of coinfection with HIV also are considered when designing optimal treatment regimens. The local health department is critical to the optimal care of a child with TB. Health departments often diagnose TB in children, identify an infectious source case and administer appropriate treatment, and provide directly observed therapy. Brief descriptions of the investigations undertaken by health departments are given in Table 4.
Chemotherapy for active TB is directed toward eradication of tubercle bacilli. Treatment goals include the prevention of multidrug-resistant TB and the complications of TB. Antimycobacterial agents ideally should be bactericidal and kill both intracellular (pyrazinamide) and extracellular (isoniazid and rif ampin) populations of M. tuberculosis. Three or more drugs are used empirically for initial therapy and later adjusted based on susceptibility patterns. Therapy should be provided for prolonged periods and given via directly observed therapy whenever possible.
Recommended First-line Drags for Pediatrie Tuberculosis
Recommended Second-line Drugs for Pediatrie Tuberculosis
During the past two decades, studies have evaluated shorter multidrug regimens that take into account local resistance patterns. Short-course studies show the successful use of isoniazid, rifampin, and pyrazinamide in both adults and children for pansusceptible TB (Tables 5 and 6). Patients initially receive all three drugs for 2 months and pyrazinamide is discontinued after 2 months. Isoniazid and rifampin are continued for 4 more months. After 2 to 4 weeks of daily therapy, intermittent therapy (ie, twice or thrice weekly) is as effective as daily therapy if provided by directly observed therapy to ensure adherence. Some physicians are more comfortable with thrice-weekly therapy for very young children because of the high rates of emesis after drug ingestion. Isoniazid tablets, rifampin capsules, or pyrazinamide tablets can be crushed and added to juice or foods.
More recent recommendations for empiric therapy include isoniazid, rifampin, and pyrazinamide plus ethambutol in locales with high rates (> 4%) of resistance to isoniazid. Once susceptibilities are known, ethambutol can be discontinued if the organism is pansusceptible. However, if the susceptibility of the source case is unknown and the child's cultures remain negative, four drugs (isoniazid, rifampin, pyrazinamide, and ethambutol) are recommended for at least 6 months. Unfortunately, fewer than 50% of gastric aspirate cultures are positive, and source cases are not always identified because many of these individuals reside abroad. Many children with TB are therefore treated empirically.
Some physicians are concerned about possible optic toxicity due to ethambutol, especially in young children who are unable to report changes in visual acuity or color vision. Others are less concerned if lower dosages (eg, 15 mg/kg/d) are used. Children receiving ethambutol should be taught to distinguish colors as soon as possible. Young children can use toys with musical buttons to identify colors.
Isoniazid-resistant TB can be treated with rifampin, pyrazinamide, and ethambutol, given for at least 6 months. Treatment of multidrug-resistant TB is guided by the susceptibilities of the source case and generally involves at least three secondline drugs used for 18 to 24 months (Tables 5 and 6). Such treatment should be in conjunction with an expert to ensure selection of the optimal regimen and should be administered only by directly observed therapy to ensure adherence.
Vitamin B6 (pyridoxine) is recommended for adolescents or adults with diets low in milk or meat to prevent peripheral neuropathy. The dose is 25 or 50 mg/ d given as a single dose. It should also be used for HIV-positive individuals and those with malnutrition. Vitamin B6 is recommended for all regimens that contain ethionamide and cycloserine. Higher dosages (50 to 100 mg) are given with cycloserine.
There is no need to obtain routine blood tests for chemistries, liver function, or hematologic studies in children with active TB because the treatment regimens are well tolerated. However, if a child has clinical hepatitis (eg, icterus, hepatomegaly, acholic stools, or dark urine), treatment should be stopped and liver function tests should be obtained immediately.
Coiniection With HIV
Although coinfection with HIV and TB is less common in children than in adults, all children with active TB should be screened for HIV. The treatment regimens for HIV-positive children are similar to those for HIV-negative children. However, more prolonged treatment courses should be considered for patients with a delayed clinical response. Treatment of such children should be done in consultation with an expert because of interactions between protease inhibitors and rifampin related to metabolism by the cytochrome P450 system.
Contacts of infectious source cases should be evaluated by TST and chest radiographs as indicated. However, an additional therapeutic intervention termed "window prophylaxis" is recommended for children 5 years or younger who were exposed to an infected adult. The rationale for window prophylaxis is that (1) young children are at increased risk for active TB when infected and early initiation of preventive treatment (prior to TST conversion) may prevent progression to active TB; (2) TST conversion may take 6 to 12 weeks and can delay appropriate therapy; and (3) if the infectious source case and young child reside in the same household, it is possible to provide directly observed therapy to the source case and directly observed prophylactic therapy to the child.
In general, regimens for pulmonary TB are effective for extrapulmonary TB. The duration of treatment for lymphadenitis is similar to that for pulmonary TB. However, central nervous system, disseminated, or skeletal diseases generally require longer courses of therapy. Patients with suspected extrapulmonary TB should also be evaluated for pulmonary TB. Initial treatment for tuberculous meningitis consists of isoniazid, rifampin, pyrazinamide, and ethambutol for 2 months. Because ethambutol penetrates the central nervous system only when the meninges are inflamed, ethambutol can be discontinued after 2 months. Isoniazid, rifampin, and pyrazinamide are continued for at least 10 more months. Such children should be observed for signs of hydrocephalus and increased intracranial pressure.
There are several indications for the use of steroids as an adjuvant in the treatment of TB. During treatment of tuberculous meningitis, steroids are indicated for changes in mental status, including confusion, stupor, obtundation, focal deficits, hydrocephalus, or cerebral edema. Steroids may reduce endobronchial lesions causing airway obstruction. Initial doses are 1 mg/kg of prednisone daily. Following improvement, steroids are tapered to complete a course of 4 to 6 weeks.
Children infected with pansusceptible or isoniazid-resistant M. tuberculosis who complete 6 months of treatment have a very low rate of relapse (< 5%), and have an excellent short- and long-term prognosis. They can safely be discharged from the clinic. However, children who do not complete treatment are at increased risk of disease progression, reactivation, or both later in life. All discharge instructions should review signs and symptoms of TB and parents should be instructed to return to the clinic should such signs and symptoms develop.
Children with multidrug-resistant TB or whose therapy did not include rifampin are thought to be at greater risk of posrtreatment relapse and therefore require close follow-up. Some physicians recommend clinical monitoring and twice annual chest radiographs for 24 months after the completion of therapy. Currently, the long-term prognosis for this group of patients appears good despite the higher relapse rate.
Latent Tuberculosis Infection
Most children with TB do not have active TB, but have latent TB infection, also termed class II by the WHO (Table 3). Patients with latent TB infection are asymptomatic, have positive results on a TST and normal results on chest x-ray, and are presumed to be infected with low numbers of viable tubercle bacilli that are dormant.
Recent studies address risk factors for latent TB infection in children and provide valuable insight into targeted TST based on an individual child's risk factors, rather than community risk factors. Targeted testing is cost-effective and improves the positive predictive value of a positive result on TST. In a case-control study conducted in northern Manhattan, contact with an adult with active TB, foreign birth, foreign travel, and a relative with a positive result on TST were risk factors for latent TB infection in children 1 to 5 years old.24 Similarly, Ozuah et al. found that children in the Bronx with a history of contact with a case of TB, birth in or travel to an endemic area, or contact with a high-risk adult (defined as an adult with HIV infection, homelessness, incarceration, or illicit drug use) were more likely to have a positive result on TST.25 Likewise, Froehlìch et al. found that children in California could be screened for TB using a standardized/ validated questionnaire that identified children who should undergo TST.26 They found that the following were associated with latent TB infection in children: a household member with TB or latent TB infection, bacille Calmette-Guérin immunization, birth in or travel to endemic areas, or parents who were Asian or Hispanic. All three studies confirm that contact investigation of an infectious adult is an effective strategy to detect children with latent TB infection. Finally, internationally adopted children are also at risk for latent TB infection.27 Thus, the optimal screening strategy is a risk factor assessment and then targeted TST.
Screening the family members of children with latent TB infection to detect active TB (Table 4) is not worthwhile; among 659 close adult and adolescent contacts of 187 children with latent TB infection, no cases of TB were found, although 32% of the contacts had a TST result of 10 mm or greater.28
The definition of a positive result on TST varies with risk factors. A TST result of 5 mm or greater is considered positive if the child is immunocompromised, has had exposure to someone with TB, or has a chest x-ray showing a calcifíed granuloma. A TST result of 10 mm or greater is considered positive if the child is 4 years of age or younger or has risk factors for latent TB infection (eg, is foreign born or has traveled to a country with high case rates of TB). A TST result of 15 mm or greater is positive if none of the above are true.
The risk for active TB once infected is inversely proportional to age and is increased by underlying conditions (Table 2). Thus, anti-tuberculous chemotherapy is used to kill all viable bacilli. As with treatment for active TB, therapy must be given for prolonged courses. In the absence of a known infectious source case, isoniazid is given for 9 months and if the infectious adult source case is known to have isoniazid-resistant TB or the patient cannot tolerate isoniazid, rifampin can be given for 6 months. There are currently no pediatrie data to support shorter 2-month courses with rifampin and pyrazinamide such as have been studied in adults. Recent reports of this shorter two-drug regimen in adults indicate an association with an increased risk of hepatotoxicity.29 Whenever possible, children should receive directly observed prophylactic therapy provided by the health department or school-based clinics. Only directly observed prophylactic therapy with isoniazid can be intermittent (eg, twice or thrice weekly).
Treatment of latent TB infection following exposure to multidrug-resistant TB should consist of a regimen that is guided by the susceptibility of the source case. Such treatment should be provided in consultation with an expert (ideally within the health department) who is familiar with die treatment of multidrug-resistant TB and with knowledge of the susceptibility of the source case. Such collaborative effort with the health department facilitates directly observed prophylactic therapy, especially if the source case is already receiving directly observed therapy and lives in the same household with the infected child. Most experts recommend that directly observed prophylactic therapy be provided to all patients receiving second-line drugs to prevent the further emergence of resistance related to nonadherence to recommended therapy. There are currently no long-term studies of the efficacy of therapy for latent TB infection caused by multidrug-resistant TB. Some experts do not recommend preventive therapy, especially for children older than 5 years, because of concern about toxicities, unproven efficacy, uncertain dosing regimens, and possible emergence of further resistance.
TB continues to be a major health problem in the United States. Targeted screening of children at high risk for latent TB infection is a cost-effective strategy. Complete laboratory and radiographie evaluation of children with suspected TB or latent TB infection is critical to ensure the appropriate diagnosis. Assessment of the close contacts of patients with TB by the health department and providing therapy for them (including treatment, preventive therapy, or window prophylaxis), if required, are crucial elements of TB control programs in the United States.
1 Doli P, Raviglione M, Kochi A. Global tuberculosis incidence and mortality during 1990-2000. Bull World Health Organ. 1994;72:213-220.
2. Centers for Disease Control and Prevention. Tuberculosis morbidity: United States. MMWR. 1998;47:253-257.
3. Styblo K. Recent advances in epidemiological research in tuberculosis. Advances in Tuberculosis Research. 1980;20:163.
4. Stead W. Tuberculosis among elderly persons: an outbreak in a nursing Home. Ann Intern Med. 1981 $4:606-610.
5. Cauthen G, Dooly S, Onorato I. Transmission of Mycobacterium tuberculosis from tuberculosis patients with HIV infection or AIDS. Am ] Epidemial. 19%;144:69-77.
6. Miller F, Seale R, Taylor M. Tuberculosis in Children. Boston: Little Brown; 1963.
7. Abadco D, Steiner P. Gastric lavage is better than bronchoalveolar lavage for isolation of Mycobacterium tuberculosis in childhood tuberculosis. Pediatr Infect Dis J. 1992;11:735-738.
8. Middlebrook G, Reggiardo Z, Tiget W. Automated radiometric detection of growth of Mycobacterium tuberculosis in selective media. American Review of Respiratory Diseases. 1977;115:1066-1069.
9. Lebrun L, Espinasse F, Poveda J. Evaluation of nonradioactive DNA probes for identification of mycobacteria. J Clin Microbio/. 1992^0:2476-2478.
10. Ehlers B, Ignatius R, Regnath T, et al. Diagnosis extrapulmonary tuberculosis by Gen-Probe amplified Mycobacterium tuberculosis direct test. / CKn Microbiol. 1996; 34:2275-2279.
11. Devallois A, Legrand E, Rastogi N. Evaluation of Amplicor MTB test as adjunct to smears and culture for direct detection of Mycobacterium tuberculosis in the French Caribbean. / CIm Microbiol. 1996;34:1065-1068.
12. Moore D, Curry J. Detection and identification of Mycobacterium tuberculosis directly from sputum sediments by Amplicor PCR. J Clin Microbiol. 1995;33:26862691.
13. Pfyffer G, Kissling P, Wirth R, et al. Direct detection of Mycobacterium tuberculosis complex in respiratory specimens by a target-amplified test system. / CKn Microbiol. 1994;32:918-923.
14. Neu N, Siaman L, San Gabriel P, et al. Diagnosis of pediatrie tuberculosis in the modern era. Pediatr Infect Dis J. 1999;18:122-126.
15. Centers for Disease Control and Prevention. Core Curriculum on Tuberculosis. Atlanta: Centers for Disease Control and Prevention; 1994.
16. Fujiwara P, ed. Clinical Policies and Protocols, 3rd ed. New York: Bureau of TB Control, New York City Department of Health; 1999.
17. Centers for Disease Control and Prevention. Case definitions for public health surveillance. MMWR. 1990;39:3940.
18. Sinder D, Reider H, Comb D, et al. Tuberculosis in children. Pediatr Infect Dis }. 1987;7:271-278.
19. Vallejo J, Ong L, Starke J. Clinical features, diagnosis, and treatment of tuberculosis in infants. Pediatrics. 1994;94:1-7.
20. Starke J, Taylor-Watt K. Tuberculosis in the pediatrie population of Houston. Pediatrics. 1989;84:28-35.
21. American Academy of Pediatrics. Tuberculosis. In: Pickering L, ed. 2000 Red Book: Report of the Committee on Infectious Diseases, 25th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2000:593-613.
22. Zar H, Tannenbaum E, Apolles P, et al. Sputum induction for the diagnosis of pulmonary tuberculosis in infants and young children in an urban setting in South Africa. Arch Dis Child. 2000;82:305-308.
23. Alvarez S, McCabe W. Extrapulmonary tuberculosis revisited: a review of experience at Boston City and other hospitals. Medicine. 1984;63:25-54.
24. Saiman L, San Gabriel P, Schulte J, Vargas MP, Kenyon T, Onorato I. Risk factors for latent tuberculosis infection among children in New York City. Pediatrics. 2001;107: 999-1003.
25. Ozuah PO, Ozuah TP, Stein REK, Burton W, Mulvihill M. Evaluation of a risk assessment questionnaire used to target tuberculin skin testing in children. JAMA. 2001; 285:451-453.
26. Froehlich H, Ackerson LM, Morozumi PA, the Pediatrie Tuberculosis Study Group of Kaiser Permanente, Northern California. Targeted testing of children for tuberculosis: validation of a risk assessment questionnaire. Pediatrics. 2001;107:e54. Available at www. pediatrics.org / cgi / content / full / 107 / 4 / e54.
27. Saiman L, Aronson J, Zhou J, et al. Prevalence of infectious diseases among internationally adopted children. Pediatrics. 2001;! 08:608-612.
28. Soren K. Saiman L, Irigoyen M, McMahon D, GomezDuarte C, Levison M. Is it productive to screen household members of children with positive tuberculin skin tests for tuberculosis? Pediatr Infect Dis J. 1999;! 8:949-955.
29. Centers for Disease Control and Prevention. Fatal and severe liver injuries associated with rifampin and pyrazinamide for latent tuberculosis infection, and revisions in American Thoracic Society /CDC recommendations, US 2001. MMWR. 2001;50:733-735.
Adults at Risk for Tuberculosis
Factors That Increase the Risk of Progression to Active Tuberculosis In Children
Modified International Classification of Tuberculosis
Investigations Conducted by Health Departments to Diagnose Tuberculosis
Recommended First-line Drags for Pediatrie Tuberculosis
Recommended Second-line Drugs for Pediatrie Tuberculosis