While numbers of tuberculosis (TB) cases in both adults and children have decreased dramatically within the United States during the past decade, TB remains a global crisis. One-third of the world's population is infected with Mycobacterium tuberculosis.
In 1989, the World Health Organization (WHO) estimated that approximately 1.3 million cases of TB in children younger than 15 occur each year, resulting in 450,000 deaths.2 More recent age-specific estimates based on reported numbers of smear-positive TB cases put the total global burden of childhood TB at about 900,000 cases among children younger than 15 in 2000.3 The vast majority of these TB cases - more than 90% - occur in the developing world.
Estimating the burden of TB disease in children is complicated by several factors: the difficulty in establishing a definitive diagnosis, the frequency of extrapulmonary disease in young children, the absence of a standard case definition, and the lower public health priority given to childhood TB versus that given to adult TB.3 Currently, WHO global reporting on TB only includes age breakdowns for smear positive TB cases, which represents at best a small subset of children with TB (studies report only 8% or less of children with TB are sputum-smear positive).4
The internationally accepted TB control strategy, DOTS - formally referred to as Directly Observed Therapy, Short Course - targets TB cases with sputum smears positive for acid-fast bacilli as the highest priority for diagnosis and treatment. These cases are the most infectious and therefore are responsible for the majority of transmission in the community. However, because children with TB are rarely sputum-smear positive, TB control programs may place less emphasis on diagnosing and treating TB in children.
Even so, TB is an important cause of morbidity and mortality in children worldwide. In industrialized countries, TB in children represents only 5% to 10% of annual TB cases. However, in resource-poor countries, this percent is as high as 20% to 25 %.3 Children with TB are sentinel cases; they represent recent transmission of TB in the community. Moreover, children younger than 5 are more likely to develop more severe forms of TB such as miliary (disseminated) TB or TB meningitis, which both are associated with higher mortality rates. In addition, children who recover from primary TB disease are the reservoir of future active, and often infectious, cases.
Estimated case rates of TB among children in the 22 highest TB-burden countries - where more than 80% of all TB cases occur - are as high as 200 per 100,000 population for some countries3 and more than 1,000 per 100,000 population in omers.1 Standardized case definitions and more inclusive global reporting of TB cases in children, for both smear-negative and extrapulmonary TB, will improve current estimates of the TB burden among children.
Fortunately, data suggest the implementation of the DOTS strategy over time will have an effect on childhood TB rates. In Beijing, China, where DOTS has been implemented effectively since 1978, iverakk decreases in the prevalence and incidence of TB cases have been accompanied by the virtual disappearance of TB meningitis cases.5
BCG VACCINE AND TB IN CHILDREN
The bacille Calmette-Guerin (BCG) vaccine, a live bacterial vaccine first used in 1921, is one of the most widely used vaccines worldwide, with about 100 million children receiving BCG vaccine each year.6 Determination of its efficacy in preventing various forms of TB has been complicated and has often yielded variable and inconsistent results. Reported measurements of the protective efficacy of BCG vaccine against pulmonary TB have ranged from 0% to 80%, with the lower range of efficacy often seen in tropical regions, where the vaccine is needed most.3,6
In a meta-analysis of efficacy studies, BCG vaccine was found to have an overall 5 1 % protective effect against all forms of TB, a 64% protective effect against TB meningitis, and a 78% protective effect against disseminated TB.7 Due to its protective effect against these fatal forms of childhood TB, the WHO recommends BCG vaccination in high TB-burden countries as soon as possible after birth.
While BCG revaccination or booster inoculations are given in some countries, no data support this practice, even among individuals who remain negative on subsequent tuberculin skin testing.6 The WHO recommends BCG vaccination of all neonates in high HIV-prevalence areas, given the high risk of developing TB and the low risk of serious adverse reactions in HIV-exposed neonates. Although the efficacy of BCG vaccination in HIV-infected infants is unknown, HIV-infected infants may receive BCG vaccine if they are asymptomatic and living in a TB-endemic area. Long-term follow-up is recommended for these children. BCG vaccination should not be given to infants or children with symptomatic HIV infection or immunosuppression due to another cause.
Despite effectiveness in protecting infants and young children from lifethreatening forms of TB, high rates of BCG vaccination have not led to a decreased incidence of TB cases worldwide because of the vaccine's limited role in preventing infectious pulmonary TB among adults.6 Until a more effective vaccine is available, children will continue to be exposed to TB and will remain at risk for developing TB disease.
PRESENTATION OF TB IN CHILDREN
The most common type of childhood TB is extrapulmonary TB, with a 3 to I ratio of extrapulmonary to pulmonary disease.8 The most frequent forms of extrapulmonary disease are TB lymphadenopathy (especially intrathoracic), TB meningitis, TB effusions, and spinal TB. A typical presentation in a child is primary TB with unilateral hilar or mediastinal lymphadenopathy without any notable parenchymal involvement on chest radiograph. When a small lung opacity accompanies this presentation, it is considered a typical primary complex.
Miliary TB is an early complication usually occurring within 3 to 6 months of primary infection. It results from hematogenous dissemination with spread to two or more organs. Acute disseminated TB - miliary TB with or without TB meningitis - is more common in children younger than 5, whereas post-primary pulmonary TB is seen more often in children older than 10.8
Symptoms of childhood TB generally are chronic and nonspecific. A study comparing symptoms between children with culture-proven TB and those with other lung diseases found no difference between the two groups with respect to weight loss, chronic cough, and duration of symptoms.9 The only clinical factors that differentiated between the two groups were history of contact with an infectious TB case and a positive tuberculin skin test (TST).
The Challenge of Diagnosing TB in Children
Confirming the diagnosis of TB in children is difficult, even in industrialized countries under the best of circumstances. In the developing world, where reliance on sputum-smear microscopy alone to diagnose TB is standard practice, even fewer pediatric TB cases are likely to be confirmed. As stated, children present with extrapulmonary TB more often than pulmonary disease, and when they do present with pulmonary TB, it is usually paucibacillary, noncavitary disease. In addition, children lack the tussive force of adults to produce adequate sputum samples.10 Sputum smear for AFB therefore is not an effective means of diagnosing TB in children.
Gastric aspiration is the most common method of obtaining sputum samples from children in industrialized countries. However, this technique rarely is performed in resource-poor countries. In addition, gastric sampling requires access to mycobacterial culture techniques to process the samples effectively - access not routinely available in many developing world settings. Even under ideal circumstances, gastric aspirate sample results are only able to confirm the diagnosis of TB in approximately 40% of cases." Due to the paucibacillary nature of childhood TB, newer nucleic acid amplification techniques are not useful methods for diagnosing TB in children either.
Figure. Chest radiograph of a 10-year-old boy who presented with fever. A small area of pneumonitis just above the right diaphram and enlarged hilar nodes were noted. On further investigation, the child's father was found to have sputum-smear-positive pulmonary TB.
Where available, the chest radiograph can be helpful in diagnosing TB in children. Typically, chest radiographs have a higher yield than either sputumsmear microscopy or mycobacterial culture of gastric aspirate contents.12 The most common finding on chest radiograph is unilateral hilar or mediastinal lymphadenopathy (Figure).
The TST is another tool used to help diagnose TB in children. A negative TST does not rule out TB, because severe malnutrition, HIV, and even disseminated TB can cause a false-negative TST result.8 Therefore, a TST is most useful when it is positive. WHO has described standard procedures for implanting a TST and defined a positive result as induration of 10 mm or greater when a child has not been vaccinated with BCG but acknowledges that 10 to 14 mm of induration in a child who has had BCG vaccination may be due to TB infection or vaccination. Because BCG vaccine is used in all parts of the developing world (with good coverage, as noted above), more often than not, the latter case is the relevant circumstance.
Regardless of the vaccination status, the TST should be interpreted in the context of the clinical features and history of TB contact. In some cases, any degree of induration is significant. It has been noted that induration in response to the TST tends to wane over time and that in areas where TB transmission is common, the TST is still a fairly good predictor of disease, even in BCG- vaccinated populations.13
Recommended Regimens for Treatment of Tuberculosis in Children8
Of note, the TST tends not to be a useful tool in immunocompromised children, such as those who are HIV-infected, because of the high rate of false-negative results caused by anergy. In addition, tuberculin often is not available in resource-poor settings, particularly rural areas where an effective cold chain is difficult to maintain.
To help overcome the diagnostic challenges posed by childhood TB, various point scoring systems, approaches, and algorithms have been developed to improve identification of children likely to have TB. In a recent review of me 16 published systems, Hesseling et i?.14 found that only seven had been evaluated (either prospectively or retrospectively), and none in controlled trials. The lack of standardized definitions and clear diagnostic guidance likely has contributed to the neglect and lack of attention to childhood TB.
Treatment of Childhood Tuberculosis
Once a diagnosis of TB is made in a child, standard treatment regimens for children, devised by the WHO, are used. Typically, because a culture result is not available, most children are started on treatment for drug-susceptible TB disease. The recommended regimens for children are shown in Table 1.
Although some concern has been expressed about the use of ethambutol in young children for whom it may not be possible to monitor visual changes, ethambutol is still used frequently during the initial phase, when four drugs are indicated. Parental and physician monitoring of any visual changes is recommended. Fortunately, this side effect rarely is seen in children. In addition, because ethambutol does not cross the blood-brain barrier, it is replaced by streptomycin in the treatment of TB meningitis.
Given the documented rises in drugresistant TB, four drugs are likely to remain the recommended standard regimen for the initial phase in cases with drug resistance. Ethambutol may be discontinued when a culture result is available and drug-susceptibility testing shows a fully susceptible organism, or when the definitive source case is known to have drug-susceptible disease. However, drug-susceptibility testing often is not available nor performed routinely in most resource-poor settings. Success rates of more than 95% have been achieved in children with pulmonary or less severe forms of extrapulmonary TB.8
It is known that children generally tolerate anti-TB medications well, but few data are available on whether they successfully complete treatment. The WHO's annual report "Global Tuberculosis Control: Surveillance, Planning, Financing" contains information on treatment outcomes only for cohorts of sputum-smear-positive patients, which aree unlikely to contain many childhood cases.15
A recent study in Malawi showed that, of 2,739 childhood TB cases (all forms), only 45% completed treatment, while 35% either transferred out (without any follow-up information) or defaulted from treatment.16 Seventeen percent died during treatment, with a higher mortality rate noted among those younger than 5. Additional research is needed to help discern if these unfavorable outcomes among children are due to adherence issues or poor treatment response. In addition, it is acknowledged that methods to improve case detection, diagnosis and treatment are needed.
Although children with TB are less likely to develop resistance during treatment, due to their lower bacillary load, children can contract multidrug-resistant TB (MDR-TB) from a source case with MDR-TB. In one study comparing 285 children with isoniazid-susceptible TB and 2 1 children with isoniazid-resistant TB (including seven with MDRTB), there were no differences in clinical presentation and chest radiographs.17 In addition, recent data suggest children tolerate MDR-TB treatment well and can have favorable outcomes.18 However, all these data are limited by small sample sizes.
While in-depth contact investigations usually are not performed in developing countries, the WHO recommends active tracing of child household contacts of all smear-positive TB cases. Evaluation of these children should include a history and physical exam, as well as a chest x-ray, TST, and HIV test if available. Any children found to have active disease should be started on appropriate therapy. When active disease has been ruled out, preventive therapy, also now referred to as "treatment for latent TB infection," should be initiated.
The International Union Against Tuberculosis and Lung Diseases recommends contact tracing be limited to all children younger than 5 who live in the same household as a sputum-smearpositive TB patient.19 Following a clinical exam to rule out active disease, all those found to be healthy would be given a course of preventive therapy with isoniazid. The rationale for this recommendation is that children younger than 5 are most likely to have had prolonged exposure in the home and, therefore, most likely to be infected, most likely to progress to active disease if infected, at greatest risk for developing severe forms of TB, at low risk of side effects from the medicine, and, even if active disease is present, at low risk of developing drug resistance, due to the paucibacillary nature of pediatric TB.19 In addition, the source case often can be enlisted to administer the therapy to the child. This recommendation also takes into account the realities of TB-control services in rural or peripheral sites, where there may be no access to tuberculin or even chest radiography. Currently, for children older than 5 in whom active disease has been ruled out, the WHO recommends only clinical followup.
EFFECTS OF HIV ON CHILDHOOD TB
It is estimated that between 11% and 64% of children with TB in sub-Saharan Africa and elsewhere also are infected with HIV.3 Co-infection with HIV further complicates the process of diagnosing TB in children. Routine diagnostic tests such as TST and chest radiograph have been found to be less sensitive among HIV-infected children with TB.3 Mycobacterial culture among HIVinfected children has been found to have a lower yield than with HIV-uninfected children.20
HFV infection also has been associated with a higher mortality rate from TB. A study in Ethiopia reported that TB mortality was six times higher in HIVinfected children.20 In addition, HIV infection has been associated with poorer treatment adherence due to parent illness or death and poorer treatment outcomes due to immune suppression.16 Children with TB who are co-infected with HTV also are reported to have higher relapse rates.3·21 Extending the duration of TB treatment from 6 months to 9 to 12 months should be considered for HIV-infected children, even when rifampin is used throughout the treatment course.21
HIV-positive children also are likely to have multiple pulmonary pathogens such as bacterial or viral pneumonia related to their HIV infection. In some cases, children may be diagnosed with TB when, in fact, they have another HIV-related Jung disease. They may also have several causes of lung disease at the same time. Diagnosis of these other HIV-associated lung diseases is difficult. Because of this complex picture, both over- and under-diagnosis of TB among HIV-infected children is possible.3
PROMISES FORTHE FUTURE
While focusing TB control efforts on sputum-smear-positive adult cases makes sense to halt ongoing transmission, it is a long-term strategy for controlling TB in children. While waiting to see the benefits of this approach, millions of children will become ill with TB, and many will die as a result. In the meantime, childhood TB must be recognized as a significant cause of morbidity and mortality, and more immediate action should be implemented.
International partners in TB control, such as the WHO, the International Union Against Tuberculosis and Lung Diseases, the International Pediatric Association, the Centers for Disease Control and Prevention, and the National Institutes of Health, are working together to address childhood TB worldwide. Creation of a pediatric TB subworking group under the WHO's Stop TB Partnership is planned to promote and coordinate research and program initiatives.4 Activities will also include efforts to improve childhood TB surveillance and creation of a database for surveillance, clinical and research data, and ongoing work.
Clearly, new diagnostic tools are needed, as are shorter treatment regimens and a more effective vaccine. Progress in each of these areas has been made through the contributions of various global initiatives and the Stop TB Partnership Working Groups on New Diagnostics, New Drugs, and Vaccine Development and their partners. For example, the TB Diagnostics Initiative has supported commercial development of new diagnostics, the Global Alliance for TB Drug Development has ensured swift movement of promising compounds through the development and approval pipelines, and two new vaccine candidates have entered phase I clinical trials.22
Tuberculosis remains an important cause of childhood morbidity and mortality in many areas of the developing world. Published data on the epidemiology of TB in children are scarce, and research efforts focused on TB in children are too few. The diagnosis of TB in children is difficult, and too many children are not completing their prescribed courses of treatment. Controlling TB in all populations requires that attention be devoted to the issues of accurate surveillance and adequate diagnosis and treatment of TB in children.
1. Donald PR. Childhood tuberculosis: the hidden epidemic, lnt J Tuberc Lung Dis. 2004: 8(5):627-629.
2. World Health Organization. Childhood Tuberculosis and BCG vaccine: Expanded Program on Immunization Update Supplement. Geneva, Switzerland: WHO; 1989.
3. Nelson LJ, Wells CD. Global epidemiology of childhood tuberculosis. Int J Tuberc Lung Dis. 2004;8(5):636-647.
4. Wells CD, Nelson LJ. New international efforts in childhood tuberculosis: proceedings from the 2002 Workshop on Childhood Tuberculosis, Montreal, Canada, 6-7 October 2002. Int J Tuberc Lung Dis. 2004;8(5):630-635.
5. Enarson DA. Children and the global tuberculosis situation. Paediatr Respir Rev. 2004;5 Suppl A:S 143- 145.
6. World Health Organization. BCG vaccine. WHO position paper. WkIy Epidemiol Ree. 2004;79(4):27-38.
7. Colditz GA, Brewer TF, Berkey CS. Efficacy of BCG vaccine in the prevention of tuberculosis. Meta-analysis of the published literature. JAMA. 1 994:27 1 (9):698-702.
8. Treatment of tuberculosis: guidelines for national programmes. Geneva, Switzerland: World Health Organization; 2003. Document WHO/CDS/TB 2003.313.
9. Schaaf HS, Beyers N, Gie PR, et al. Respiratory tuberculosis in childhood: the diagnostic value of clinical features and special investigations. Pediatr Inf Dis J. 1995; 14(3): 189-94.
10. Pereira L. Tuberculosis: role of etiologic diagnosis and tuberculin skin test. Pediatr Pulmonol Suppl. 2004;26:240-2.
11. Starke JR. Pediatric tuberculosis: time for a new approach. Tuberculosis (Edinb). 2003; 83(1-3):208-212.
12. Gie RP, Beyers N, Schaaf HS, Goussard P. The challenge of diagnosing tuberculosis in children: a perspective from a high incidence area. Paediatr Respir Rev. 2004;5 Suppl A:S147-149.
13. Enarson DA. Use of the tuberculin skin test in children. Paediatr Respir Rev. 2004;5 Suppl A:S135-137.
14. Hesseling AC, Schaaf HS, Gie RP, Starke JR, Beyers N. A critical review of diagnostic approaches used in the diagnosis of childhood tuberculosis. Int J Tuberc Lung Dis. 2002; 6(12):1038-1045.
15. WHO. Global Tuberculosis Control: Surveillance, Planning, Financing. WHO Report 2004. Geneva, Switzerland: WHO; 2004.
16. Harries AD, Hargreaves NJ, Graham MS, et al. Childhood tuberculosis in Malawi: nationwide case-finding and treatment outcomes. Int J Tuberc Lung Dis. 2002;6 (5):424-431.
17. Schaaf HS, Gie RP, Beyers N, et al. Primary drug-resistant tuberculosis in children. Int J Tuberc Lung Dis. 2000;4(12):1 149-1155.
18. Mukherjee JS, Joseph JK, Rich ML, et al. Clinical and programmatic considerations in the treatment of MDR-TB in children: a series of 16 patients from Lima, Peru, hit J Tuberc Lung Dis. 2003;7(7):637-644.
19. Reider HL. Contacts of tuberculosis patients in high-incidence countries. Int J Tuberc Lung Dis. 2003;7(12 Suppl 3):S333-336.
20. Berggren Palme I, Gudetta B, Bruchfeld J, Eriksson M, Giesecke J. Detection of Mycobacterium tuberculosis in gastric aspirate and sputum collected from Ethiopian HIV-positive and HIV-negative children in a mixed in- and outpatient setting. Acta Paediatr. 2004;93(3):31 1 -5.
21. Cotton MF, Schaaf HS, Hesseling AC, et al. HIV and childhood tuberculosis: the way forward. Int J Tuberc Lung Dis. 2004:8 (5):675-82.
22. Progress report on the global plan to stop tuberculosis. Geneva. Switzerland: World Health Organization; 2004. Document WHO/HTM/STB/2004.29.
Recommended Regimens for Treatment of Tuberculosis in Children8