Tuberculosis - Persistent Threat to Human Health
http://www.100md.com
《美国医学杂志》
Jaslok Hospital and Medical Research Center, Mumbai, Maharashtra, India
Abstract
With the increasing incidence of tuberculosis worldwide, childhood cases now constitute 40% of the total. TB control thus has global importance. Unfortunately, control of disease is not in sight. It was always thought that adult tuberculosis is the fountainhead of childhood tuberculosis but it is being increasingly realized that it is the infection acquired during childhood that promotes reactivation of adult disease, which in turn maintains the chain of transmission.Thus childhood tuberculosis needs equal or more attention for effective control. Early detection by simple tests and ensuring treatment compliance is the goal. The small number of bacilli and inaccessible sites for bacteriological confirmation makes diagnosis of childhood tuberculosis difficult. Circumstantial evidence is often the basis of diagnosis. However, as clinical manifestations depend upon host immune response and virulence of tubercle bacilli, there is no typical clinical presentation. A large number of infected children may remain asymptomatic, undiagnosed and untreated. Conventional tests such as tuberculin test and radiology are not fully dependable and newer tests have limitations. Poor patient treatment compliance contributes to failure of a tuberculosis control program and leads to drug resistance. To combat this, direct observed treatment (DOTS) has been unanimously recommended in treatment of tuberculosis. DOTS is however estimated to be used in less than 40% of new cases. Misconceptions threaten to undermine continued success in tuberculosis control. TB control is essentially a management problem. Greater accountability of governments, donors and providers is essential.
Keywords: Childhood TB; India; DOTS
Epidemiology
There is clear evidence that the worldwide incidence of tuberculosis is increasing. It is estimated that nearly 1 billion people will be newly infected with tuberculosis between 2000 and 2020. Two hundred million people will develop the disease, and 35 million will die from tuberculosis. Eighty percent of the global burden of tuberculosis is carried by 23 countries, nearly all of which are tropical.[1]
TB has declined dramatically in industrialized countries. However, their populations will become increasingly vulnerable. In USA, TB had been declining at the rate of 6% each year. In 1985, the decline stopped and thereafter cases have been increasing again.[2]
It is estimated that the global incidence is 1.3 million TB cases with a mortality rate of 450,000 per year in children < 15 years of age. Developing countries carry the highest burden with > 90% of cases and > 95% of deaths. Children <5 years of age are most affected, followed by a sharp drop until the mid-teens. The lifetime risk for developing TB disease after infection is 43% in infants, 24% in children aged 1 to 5 years, and 15% in adolescents, compared to immunocompetent adults, who have a lifetime risk of 5% to 10%. Younger children also experience more severe disease such as meningitis or disseminated disease.[3] With increasing incidence of tuberculosis in a population, the percentage of the tuberculosis caseload due to the childhood population has been estimated to increase exponentially reaching 40% of the total cases.[4] TB control thus has global importance.
Control not in sight
The global incidence rate of tuberculosis was growing at approximately 1.1% per year, and the number of cases at 2.4% per year. Early detection by simple tests and compliant treatment is the goal. Based on case reports and WHO estimates, 18 countries had reached the targets for case detection and cure by the end of 2002. However, Vietnam was the only high-burden country among them.
Global average detection rate has been around 46% and has not changed much between 1995 and 2002. The global, smear-positive case detection rate was 37% in 2002, over half way to the 70% target, and rising more quickly than at any time since 1995. Based on recent trends, we expect the case detection rate to be about 50% by 2005. Smear-positive case detection by DOTS program could be increased from 37% to 50% simply by ensuring that the diagnosis and treatment of known TB cases in South-East Asia conforms to DOTS standards.[5]
CHILDHOOD TUBERCULOSIS - CONTINUING CHALLENGES
Accurate data on childhood tuberculosis is not available due to inherent difficulties in diagnosis and also it being given low priority. It was always thought that adult tuberculosis is the fountainhead of childhood tuberculosis but it is being increasingly realized that it is the infection acquired during childhood that promotes reactivation of disease in adults, which in turn maintains the chain of transmission. It has been debated whether adult tuberculosis represents reactivation of childhood infection or exogenous reinfection. It is now well recognized that contribution of reactivation to the total load of adult disease varies between 16% in low incidence area to 80% in high incidence area.[6],[7] Thus childhood tuberculosis needs equal or more attention for effective control.
There exists a pyramid of childhood tuberculosis. The base of the pyramid consists of the large majority of infected children who continue to be asymptomatic or present non-specific symptoms and hence remain undiagnosed and untreated. They form the bulk of the pool of chronic disease in adults. In the middle of the pyramid are symptomatic patients who even if diagnosed are either untreated or inadequately treated. At the apex of the pyramid are just 1-6% of hospitalized patients, who present with serious forms of disease and face high mortality and morbidity.
Immunological considerations
Mycobacterium tuberculosis is one of the most successful pathogens of mankind, infecting one-third of the global population and claiming over two million lives every year. The ability of the bacteria to persist in the form of a long-term asymptomatic infection, referred to as latent tuberculosis, is central to the biology of the disease and can subsequently reactivate to cause active disease. The latent state of infection poses a major obstacle to eradicating tuberculosis. In latent tuberculosis, the host immune response is capable of controlling the infection and yet falls short of eradicating the pathogen. Thus host immune response contributes to the maintenance of latent tuberculous infection. However, as this immune response is sufficient to prevent disease in the majority of infected persons, it provides compelling evidence that immunity to tuberculosis does exist. There is undoubtedly a genetic component of host resistance to M. tuberculosis .
A functional equilibrium between the efficacy of cell-mediated immunity (necessary for eradication of the pathogen) and tissue-destructive delayed-type hypersensitivity reaction is observed in infected individuals. Interleukin -12 (IL-12) and gamma-interferon (IFN-g), as strong activators of the Th1-driven immune response, are considered crucial for successful eradication of mycobacteria. The functional predominance of humoral antibodies against tuberculosis is a distinctive immune response in childhood and is regarded as the risk factor associated with poor clinical outcome.[8] It is shown that antibody response predominates at the onset of disease and peaks further as disease advances while T-cell immune response is at the lowest. As disease is controlled, T-cell response recovers and antibodies go down.
Immunosuppression in Tuberculosis
Certain components of the bacterial cell wall such as lipoarabinomannan (LAM) and certain secretary proteins (30 kDa antigen or 58 kDa antigen) induce cytokine production by macrophages and contribute to systemic immune suppression and local pathology. Also IL-10 and TGF-beta produced by monocytes are suppressive factors.[9] Thus the bacterial population itself causes immune suppression and this is reflected in clinical practice in the form of a negative tuberculin test in serious and disseminated form of tuberculosis. Such a negative tuberculin test reverts to a positive reaction within 2-3 months of successful treatment of disease and may serve as retrospective confirmation of the correct diagnosis.
Co-infection with HIV and TB
Tuberculosis has re-emerged in high proportions with spread of HIV infection. HIV infection causes about a tenfold increase in TB incidence with a much higher risk in patients who have clinically advanced disease. In infants and young children, HIV infection occurs first, commonly through maternal transmission. As immune suppression supervenes, opportunistic infections such as fungi and pneumocystis carini may manifest. Tuberculosis may occur as a co-infection if the infant is exposed by chance and may manifest as more severe disease. In adults, occult tubercular infection often exists prior to exposure to HIV infection. As immune suppression occurs, occult tuberculosis becomes manifest and thus tuberculosis can be a marker of HIV infection in adults but not so in children.
Helminthic Infections and TB
A good cellular or Th1 immune response is necessary to contain M tuberculosis infection. Helminthic infections induce a Th2 immune response that suppresses the Th1 response. It has been postulated that helminthic infections predispose patients to the development of TB. Helminthic infections, being endemic in developing countries, where 95% of patients with TB live, could be an important contributing factor.[10]
Hypothetical Model Based on Immune Response
It is clear from the above-mentioned discussion that outcome of infection in tuberculosis depends upon host immune response in terms of degree of cell mediated immunity and hypersenisitivity. Such immune response is governed by multiple variable factors in relation to the host and bacteria. Age, nutrition, vaccination status and previous exposure to mycobacteria play a significant role in the ultimate host response. The number and virulence of the bacteria also decide the outcome to some extent. Drug therapy modifies the response mostly in favor of the host. A hypothetical model can be constructed for simplified understanding based on the leprosy model.
It is clear that there would be an overlap of immune response resulting in changing pathology. Decrease in protective immunity may lead to progression of local pathology as in progressive primary complex or dissemination to distant organs as in miliary TB.
Hypersensitivity response may promote destructive pathology as in cavitatory pulmonary disease or TB meningitis.
Clinical manifestations are variable
Clinical manifestations of childhood tuberculosis are also dependent upon host immune response and naturally are extremely variable. Secretion of pro-inflammatory cytokines such as TNF, IL-1 and IL-6 may contribute to symptomatology. The known consequences of TNF secretion include fever and cachexia, two prominent symptoms in severe tuberculosis. Thus no constellation of symptoms could be considered as typical of tuberculosis.
Acute-onset high fever of short duration is seen in pleural effusion. Low grade to moderate fever of long duration may be a feature of chronic or disseminated disease. Fever off and on may be the initial manifestation simulating recurrent illness.
Cough is also a variable symptom in pulmonary tuberculosis. It may be absent or mild in the initial stages and increases only in case of progressive inflammation and resultant necrosis. Hemoptysis is rare in children unlike in adults. Loss of appetite and weight is not dependable for diagnosis of tuberculosis as such symptoms are common in many childhood diseases.
Physical signs denote type of pathology and are not specific for tuberculosis. However, few pathological lesions are almost pathognomonic of tuberculosis in our setting and these include acute onset pleural effusion in a healthy older child, fibrocaseous cavitatory pulmonary disease and miliary (radiological) lesions.
Pattern of tuberculosis has changed over the years with the changing immune status of children. Cavitatory pulmonary tuberculosis was conventionally the disease of older children and so also bone tuberculosis. However with sensitization due to early exposure to mycobacteria, it may also occur in early childhood and we have seen it occasionally even in infancy. Till recently, pulmonary tuberculosis formed more than 85% of childhood tuberculosis but the advent of HIV infection has resulted in a rise of extra-pulmonary TB, which may constitute about 40% of the total burden.
In TBM, focal lesions are often hidden as long as primary disease is under control and new signs appearing thereafter may suggest that these occult lesions have now become overt. Subacute or chronic encephalopathy may manifest after withdrawal of steroids in the treatment of acute meningitis. The widespread use of BCG vaccine may have influenced the manifestations of tuberculosis in vaccinated children. Neurotuberculosis is three times more common in non-vaccinated than vaccinated children.[11] The manifestations of neurotuberculosis may be atypical and less specific in vaccinated children.
Diagnostic difficulties in childhood tuberculosis
A combination of small number of bacilli and inaccessible sites for bacteriological confirmation makes diagnosis of childhood tuberculosis rather difficult. Circumstantial evidence is often the basis of probable diagnosis. Newer tests are available but with significant limitations.
Interpretation of the tuberculin test
BCG vaccination status can be ignored while interpreting tuberculin test. More than 50% of children demonstrate negative tuberculin test after BCG vaccination. From amongst those who are positive initially, majority become negative within a year of vaccination while the remaining do so within the next 2-3 years.
Ideally 1 TU of PPD with tween 80 as a preservative should be used for testing. 5 TU without preservative may be acceptable. Use of higher strength PPD (10 TU) should be avoided, as the reaction is dose dependent. There exists a large inter-observer difference in reading the test to the tune of 15-50%. It is the induration and not just the erythema that has to be measured. Soft induration with slight edema and painless pink erythema suggests reaction to BCG vaccine while hard induration with red or purplish painful erythema denotes Koch's phenomenon. There may be formation of bullae or ulceration.
A cut off point of 10 mm is considered a result of natural infection on the basis of epidemiological studies. However tuberculin reactions follow a bell shaped curve with few children showing higher reactions even to artificial infection (BCG vaccine) and some reacting less than 10 mm to natural infection. In an immunocompromised host, a reaction of 5 mm is considered suggestive of natural infection. Initial reaction of < 10mm and increasing by 6 mm to > 10 mm within two years is considered evidence of recent infection.[12]
Bacteriology
It is the gold standard for diagnosis of tuberculosis. However, the paucibacillary nature of childhood disease and difficulty in obtaining a proper sample for testing are impediments to diagnosis. Nevertheless every attempt must be made to confirm the diagnosis by the available bacteriological tests.
It is possible to detect bacilli on smear in specimens obtained through fine needle aspiration of lymph nodes, besides the additional cytological information. Sputum may be available in older children especially in those with chronic cavitatory lesion or bronchiectesis. ZN stain is positive only if the number of bacteria is more than 10, 000 per ml of specimen. Unfortunately the yield is less than 20%. Though auromine staining increases the yield, it needs a fluorescent microscope.
Gastric aspirates are used in lieu of sputum in younger children. Early morning samples should be obtained. Initially the stomach contents are aspirated and then small amount of sterile water are injected into the stomach and then re aspirated and added to the specimen. Since gastric acidity is poorly tolerated by tubercle bacilli neutralization of the specimen should be performed immediately with 10% sodium bicarbonate or 40% anhydrous sodium phosphate. With careful attention to details and meticulous technique, tubercle bacilli can be detected in 70% of infants and 30% of older children. Broncho-alveolar lavage (BAL) is more invasive and its yield is poorer than an ideally collected gastric aspirate.[13] An Indian study reported no difference in isolation rates from gastric lavage and BAL.[14]
Conventional culture method uses LJ medium. It is less efficient and labor intensive. The yield is about 30-50%. BACTEC is a radiometric assay that is quick but not superior to a conventional culture method. Septi-check AFB system is more sensitive than other methods but to date no method has proved to be significantly superior to conventional culture. Mycobacterial growth indicator tuber system (MGIT) is a promising test still under research.
Bacterial detection in CSF and other serous fluids is less than 10%. Biochemical tests are used to differentiate TB meningitis from other forms of intracranial infections. CSF ADA values of >8 U/L has 92% specificity and 68% sensitivity for diagnosis of TBM. It has however not proved to be a satisfactory method of distinguishing different neurological infections in infancy and childhood.
Molecular diagnosis
PCR and other amplification techniques were initially thought to be very promising tools for the rapid diagnosis of tuberculosis; however several caveats remain. Contamination of samples by products of previous amplification and presence of inhibitors in the sample may lead to false positive or false negative results. PCR sensitivity show wide ranges of sensitivity between 4-80% and specificity of 80-100% and hence such tests require careful clinical appraisal and judgment.[15]
Serodiagnosis of TB
The challenge is to find an antigen that will be highly sensitive and specific. Since mycobacterial cell wall is made up of numerous proteins, lipids and carbohydrates, there are a number of antigens and epitopes to which the immune system may respond. In addition there may be sensitization due to cross-reacting antigens of environmental mycobacteria or BCG. Antibodies to 38 kDa antigen are elicited during the advanced stage of illness and to 14 kDa and 19 kDa in the early stages. Twenty percent of smear positive and large proportion of smear negative and extrapulmonary TB are seronegative and hence serology cannot be recommended. A60 IgG was positive in only 30% and 38 kDa antibody in only 50% of culture positive cases.[16]
Management issues
Poor patient compliance to anti-tuberculosis therapy is the major contributory factor to failure of the tuberculosis control program and has led increasingly to drug resistance.[17] As a way of achieving high rate of completion of treatment and preventing drug resistance, direct observed treatment has been unanimously recommended in the treatment of tuberculosis. Supervised administration of rifampicin-containing short- course chemotherapy for 6 months (DOTS) has been shown to be successful in as many as 78% of bacteriological confirmed cases compared to only a 38% success rate in non-supervised therapy. Targets set by WHO of 78% detection rate and 85% cure rate is likely to be achieved only by 2013 rather than the projected 2005, as DOTS coverage has increased at a slower rate than anticipated. HIV-infected patients with TB also respond well to short-course therapy; relapse rates are low and similar those of HIV-negative patients.
DOTS is replacing inferior treatment protocols but still is being used in fewer than 40% of estimated new TB cases. Misconceptions threaten to undermine continued success in tuberculosis control. The first misconception is that treatment observation is unnecessary. Treatment observation needs to be made more patient-friendly, but must not be abandoned. The second misconception is that health care reform will strengthen tuberculosis control. TB control is essentially a management problem. Greater accountability of governments, donors and providers is essential. A third misconception is to focus on treating multi-drug-resistant tuberculosis (MDRTB) cases without addressing the root causes of MDRTB. While it is important clinically and epidemiologically in some contexts to care optimally for patients with MDRTB, it is more important to address the causes of MDRTB and to fix these. The fourth misconception is an inordinate concern for sustainability. Delaying assistance will make implementation and sustainability in the future more difficult. Tuberculosis control is remarkably inexpensive and cost-effective, but efforts will fail unless programs have the ability to hire staff, purchase supplies, and contract for services efficiently. Critical issues for the future of tuberculosis control are sustained funding, technical rigor, and good management.[18]
In most of the areas, DOTS is a clinic based service and hence the cost to patients in developing countries is high. Several alternative models have been tried. Trained community lay supervisors have proved successful in South Africa. Using a patient nominated family member, as the drug supervisor is more convenient for patients. Other beneficial strategies include reminder cards sent to defaulters, monetary incentives offered to patients and increased supervision of TB clinic staff. It is not possible to determine from current trials whether health education by itself leads to better adherence to treatment. Even though DOTS is widely advocated as the most cost-effective means of ensuring completion of therapy, no completed trials could be found which confirm or refute this view.[19]
Problems of MDRTB
The early diagnosis of MDRTB and case management by experienced teams from the outset remains the best hope for these patients.[20] Adequately supervised and prolonged combination chemotherapy is essential, with drug choice governed mainly by quality-controlled in vitro drug susceptibility data. There is a more limited role for surgery and immunomodulating therapy, like gamma-interferon, may be a useful adjunct. Clearly the most important therapeutic modality for MDRTB is to treat drug-sensitive TB correctly in the first instance and prevent the emergence of resistant TB. Indian studies also consider DOTS as an effective measure of preventing MDRTB.[21]
For populations in which MDRTB is endemic, the outcome of the standard short-course chemotherapy regimen remains uncertain. Unacceptable failure rates have been reported and resistance to additional agents may be induced. As a consequence there have been calls for well-functioning DOTS programmes to provide additional services in areas with high rates of multidrug-resistant tuberculosis. These "DOTS-plus for MDRTB programmes" may need to modify all five elements of the DOTS strategy: the treatment may need to be individualized rather than standardized; laboratory services may need to provide facilities for on-site culture and antibiotic susceptibility testing; reliable supplies of a wide range of expensive second-line agents would have to be supplied; operational studies would be required to determine the indications for and format of the expanded programmes; financial and technical support from international organizations and western governments would be needed in addition to that obtained from local governments.[22]
Potential Approaches to Immunotherapy
In MDRTB, immunotherapy may be an adjuvant to chemotherapy. A vaccine that stimulates selectively a Th 1 response (M. vaccae) has been tried with some success.[23]
Cytokines that enhance bacillary elimination such as IL-2, IL-12 or gamma-IFN are considered for use though high cost, toxicity and large volume required for injection are impediments for its implementation. Neutralizing antibodies to immunosuppressive cytokines such IL-4, IL-10 or to cytokines that lead to immunopathology such as TNF-alpha are possible interventions.
Global emergency of tuberculosis
In view of the global emergency of tuberculosis, WHO "Stop TB" campaign has called for universal adoption of its directly observed therapy, short course strategy (DOTS). Also, through the Massive Effort Against Diseases of Poverty, several international agencies are urging the establishment of effective control programmes worldwide.[24]
References
1. WHO Global tuberculosis control. WHO Report 2001, Geneva, Switzerland: World Health Organization; 2001.
2. Kalaudt K. TB a global emergency WHO report on the TB epidemic, Gemeva, WHO 1994; 3.
3. Donald PR. Children and tuberculosis: protecting the next generation Lancet 1999; 353: 1001.
4. Donald PR. Childhood tuberculosis: out of control Curr Opin Pulm Med 2002; 8: 178-82.
5. WHO Report HTM/TB/2004: 331.
6. Bandera A, Gori A, Catozzi L, Degli Esposti A, Marchetti G, Molteni C et al. Molecular epidemiology study of exogenous reinfection in an area with a low incidence of tuberculosis. J Clin Microbiol 2001; 39 : 2213-2218.
7. Van Rie A, Warren R, Richardson M, Van Der Spuy G, Victor T et al. Exogenous reinfection as a cause of recurrent tuberculosis after curative treatment. N Engl J Med 1999; 341 : 1174-1179.
8. Flynn JL, Chan J. Immunology of tuberculosis. Annu Rev Immunol 2001; 19: 93-129.
9. Ellner JJ. Suppressor adherent cells in human tuberculosis. J Immunol 1978; 121 : 2573-2578.
10. Bentwich Z, Kalinkovich A, Weisman Z, Borkow G, Beyers N, Beyers AD. Can eradication of worms change the face of AIDS and tuberculosis Immunol Today 1999; 20 : 485-487.
11. Somu N, Vijaysekharan D, Balchandran A, Subramanyan L, Chandrabhushanam A. Tuberculous disease in a pediatric referral center - 16 years experience. Indian Pediatr 1994; 31: 1245-1249.
12. American Thoracic Society. The tuberculin skin test. Official ATS statement. Am Rev Respir 1984; 44.9
13. Abadco DL, Steiner P. Gastric lavage is better than bronchoalveolar lavage for isolation of M.TB in childhood pulmonary disease. Pediatr Infect Dis J 1992; 11 : 735-738.
14. Meenu Singh, NV. Ahsan Moosa, Lata Kumar, Meera Sharma Gastric lavage and broncho-alveolar lavage in the bacteriological diagnosis of childhood pulmonary tuberculosis. Indian Pediatr 2000; 37 : 947-951.
15. Smith KC, Starke JR. Eisencah Detection of MTB in clinical specimens from children using PCR. Pediatrics 1996; 97 : 155-160.
16. Swaminathan S, Umadevi P, Shantha S, Radhakrishnan A, Datta M Serodiagnosis of TB in children using two ELISA kits Indian J Pediatr 1999; 66 : 837-842.
17. Meenu Singh Adherence to anti-tuberculosis treatment. Indian Pediatr 1999; 36 : 1285-1286.
18. Frieden TR, Driver CR. Tuberculosis control: past 10 years and future progress. Tuberculosis (Edinb) 2003; 83(1-3) : 82-85.
19. T. Jacob John. Tuberculosis control without protection from BCG vaccine. Indian Pediatr 2000; 37 : 9-18
20. Drobniewski FA, Balabanova YM. The diagnosis and management of multiple-drug-resistant-tuberculosis at the beginning of the new millenium. Int J Infect Dis 2002; 6: S21-31.
21. Prabhakar R. Tuberculosis control in India-past, present and future. J Indian Med Assoc 2000; 98(3): 123-125.
22. Bastian I, Rigouts L, Van Deun A, Portaels F. Directly observed treatment, short-course strategy and multidrug-resistant tuberculosis: are any modifications required Bull WHO 2000; 78(2) : 38051.
23. Stanford JL, Stanford CA, Grange JM, Lan NN, Etemadi A. Does immunotherapy with heat-killed Mycobacterium vaccae offer hope for the treatment of multi-drug-resistant pulmonary tuberculosis Respir Med 2001; 95(6) : 444-447.
24. Grange JM, Zumla A. The global emergency of tuberculosis. JR Soc Health 2002; 122(2): 78-81.(Amdekar YK)
Abstract
With the increasing incidence of tuberculosis worldwide, childhood cases now constitute 40% of the total. TB control thus has global importance. Unfortunately, control of disease is not in sight. It was always thought that adult tuberculosis is the fountainhead of childhood tuberculosis but it is being increasingly realized that it is the infection acquired during childhood that promotes reactivation of adult disease, which in turn maintains the chain of transmission.Thus childhood tuberculosis needs equal or more attention for effective control. Early detection by simple tests and ensuring treatment compliance is the goal. The small number of bacilli and inaccessible sites for bacteriological confirmation makes diagnosis of childhood tuberculosis difficult. Circumstantial evidence is often the basis of diagnosis. However, as clinical manifestations depend upon host immune response and virulence of tubercle bacilli, there is no typical clinical presentation. A large number of infected children may remain asymptomatic, undiagnosed and untreated. Conventional tests such as tuberculin test and radiology are not fully dependable and newer tests have limitations. Poor patient treatment compliance contributes to failure of a tuberculosis control program and leads to drug resistance. To combat this, direct observed treatment (DOTS) has been unanimously recommended in treatment of tuberculosis. DOTS is however estimated to be used in less than 40% of new cases. Misconceptions threaten to undermine continued success in tuberculosis control. TB control is essentially a management problem. Greater accountability of governments, donors and providers is essential.
Keywords: Childhood TB; India; DOTS
Epidemiology
There is clear evidence that the worldwide incidence of tuberculosis is increasing. It is estimated that nearly 1 billion people will be newly infected with tuberculosis between 2000 and 2020. Two hundred million people will develop the disease, and 35 million will die from tuberculosis. Eighty percent of the global burden of tuberculosis is carried by 23 countries, nearly all of which are tropical.[1]
TB has declined dramatically in industrialized countries. However, their populations will become increasingly vulnerable. In USA, TB had been declining at the rate of 6% each year. In 1985, the decline stopped and thereafter cases have been increasing again.[2]
It is estimated that the global incidence is 1.3 million TB cases with a mortality rate of 450,000 per year in children < 15 years of age. Developing countries carry the highest burden with > 90% of cases and > 95% of deaths. Children <5 years of age are most affected, followed by a sharp drop until the mid-teens. The lifetime risk for developing TB disease after infection is 43% in infants, 24% in children aged 1 to 5 years, and 15% in adolescents, compared to immunocompetent adults, who have a lifetime risk of 5% to 10%. Younger children also experience more severe disease such as meningitis or disseminated disease.[3] With increasing incidence of tuberculosis in a population, the percentage of the tuberculosis caseload due to the childhood population has been estimated to increase exponentially reaching 40% of the total cases.[4] TB control thus has global importance.
Control not in sight
The global incidence rate of tuberculosis was growing at approximately 1.1% per year, and the number of cases at 2.4% per year. Early detection by simple tests and compliant treatment is the goal. Based on case reports and WHO estimates, 18 countries had reached the targets for case detection and cure by the end of 2002. However, Vietnam was the only high-burden country among them.
Global average detection rate has been around 46% and has not changed much between 1995 and 2002. The global, smear-positive case detection rate was 37% in 2002, over half way to the 70% target, and rising more quickly than at any time since 1995. Based on recent trends, we expect the case detection rate to be about 50% by 2005. Smear-positive case detection by DOTS program could be increased from 37% to 50% simply by ensuring that the diagnosis and treatment of known TB cases in South-East Asia conforms to DOTS standards.[5]
CHILDHOOD TUBERCULOSIS - CONTINUING CHALLENGES
Accurate data on childhood tuberculosis is not available due to inherent difficulties in diagnosis and also it being given low priority. It was always thought that adult tuberculosis is the fountainhead of childhood tuberculosis but it is being increasingly realized that it is the infection acquired during childhood that promotes reactivation of disease in adults, which in turn maintains the chain of transmission. It has been debated whether adult tuberculosis represents reactivation of childhood infection or exogenous reinfection. It is now well recognized that contribution of reactivation to the total load of adult disease varies between 16% in low incidence area to 80% in high incidence area.[6],[7] Thus childhood tuberculosis needs equal or more attention for effective control.
There exists a pyramid of childhood tuberculosis. The base of the pyramid consists of the large majority of infected children who continue to be asymptomatic or present non-specific symptoms and hence remain undiagnosed and untreated. They form the bulk of the pool of chronic disease in adults. In the middle of the pyramid are symptomatic patients who even if diagnosed are either untreated or inadequately treated. At the apex of the pyramid are just 1-6% of hospitalized patients, who present with serious forms of disease and face high mortality and morbidity.
Immunological considerations
Mycobacterium tuberculosis is one of the most successful pathogens of mankind, infecting one-third of the global population and claiming over two million lives every year. The ability of the bacteria to persist in the form of a long-term asymptomatic infection, referred to as latent tuberculosis, is central to the biology of the disease and can subsequently reactivate to cause active disease. The latent state of infection poses a major obstacle to eradicating tuberculosis. In latent tuberculosis, the host immune response is capable of controlling the infection and yet falls short of eradicating the pathogen. Thus host immune response contributes to the maintenance of latent tuberculous infection. However, as this immune response is sufficient to prevent disease in the majority of infected persons, it provides compelling evidence that immunity to tuberculosis does exist. There is undoubtedly a genetic component of host resistance to M. tuberculosis .
A functional equilibrium between the efficacy of cell-mediated immunity (necessary for eradication of the pathogen) and tissue-destructive delayed-type hypersensitivity reaction is observed in infected individuals. Interleukin -12 (IL-12) and gamma-interferon (IFN-g), as strong activators of the Th1-driven immune response, are considered crucial for successful eradication of mycobacteria. The functional predominance of humoral antibodies against tuberculosis is a distinctive immune response in childhood and is regarded as the risk factor associated with poor clinical outcome.[8] It is shown that antibody response predominates at the onset of disease and peaks further as disease advances while T-cell immune response is at the lowest. As disease is controlled, T-cell response recovers and antibodies go down.
Immunosuppression in Tuberculosis
Certain components of the bacterial cell wall such as lipoarabinomannan (LAM) and certain secretary proteins (30 kDa antigen or 58 kDa antigen) induce cytokine production by macrophages and contribute to systemic immune suppression and local pathology. Also IL-10 and TGF-beta produced by monocytes are suppressive factors.[9] Thus the bacterial population itself causes immune suppression and this is reflected in clinical practice in the form of a negative tuberculin test in serious and disseminated form of tuberculosis. Such a negative tuberculin test reverts to a positive reaction within 2-3 months of successful treatment of disease and may serve as retrospective confirmation of the correct diagnosis.
Co-infection with HIV and TB
Tuberculosis has re-emerged in high proportions with spread of HIV infection. HIV infection causes about a tenfold increase in TB incidence with a much higher risk in patients who have clinically advanced disease. In infants and young children, HIV infection occurs first, commonly through maternal transmission. As immune suppression supervenes, opportunistic infections such as fungi and pneumocystis carini may manifest. Tuberculosis may occur as a co-infection if the infant is exposed by chance and may manifest as more severe disease. In adults, occult tubercular infection often exists prior to exposure to HIV infection. As immune suppression occurs, occult tuberculosis becomes manifest and thus tuberculosis can be a marker of HIV infection in adults but not so in children.
Helminthic Infections and TB
A good cellular or Th1 immune response is necessary to contain M tuberculosis infection. Helminthic infections induce a Th2 immune response that suppresses the Th1 response. It has been postulated that helminthic infections predispose patients to the development of TB. Helminthic infections, being endemic in developing countries, where 95% of patients with TB live, could be an important contributing factor.[10]
Hypothetical Model Based on Immune Response
It is clear from the above-mentioned discussion that outcome of infection in tuberculosis depends upon host immune response in terms of degree of cell mediated immunity and hypersenisitivity. Such immune response is governed by multiple variable factors in relation to the host and bacteria. Age, nutrition, vaccination status and previous exposure to mycobacteria play a significant role in the ultimate host response. The number and virulence of the bacteria also decide the outcome to some extent. Drug therapy modifies the response mostly in favor of the host. A hypothetical model can be constructed for simplified understanding based on the leprosy model.
It is clear that there would be an overlap of immune response resulting in changing pathology. Decrease in protective immunity may lead to progression of local pathology as in progressive primary complex or dissemination to distant organs as in miliary TB.
Hypersensitivity response may promote destructive pathology as in cavitatory pulmonary disease or TB meningitis.
Clinical manifestations are variable
Clinical manifestations of childhood tuberculosis are also dependent upon host immune response and naturally are extremely variable. Secretion of pro-inflammatory cytokines such as TNF, IL-1 and IL-6 may contribute to symptomatology. The known consequences of TNF secretion include fever and cachexia, two prominent symptoms in severe tuberculosis. Thus no constellation of symptoms could be considered as typical of tuberculosis.
Acute-onset high fever of short duration is seen in pleural effusion. Low grade to moderate fever of long duration may be a feature of chronic or disseminated disease. Fever off and on may be the initial manifestation simulating recurrent illness.
Cough is also a variable symptom in pulmonary tuberculosis. It may be absent or mild in the initial stages and increases only in case of progressive inflammation and resultant necrosis. Hemoptysis is rare in children unlike in adults. Loss of appetite and weight is not dependable for diagnosis of tuberculosis as such symptoms are common in many childhood diseases.
Physical signs denote type of pathology and are not specific for tuberculosis. However, few pathological lesions are almost pathognomonic of tuberculosis in our setting and these include acute onset pleural effusion in a healthy older child, fibrocaseous cavitatory pulmonary disease and miliary (radiological) lesions.
Pattern of tuberculosis has changed over the years with the changing immune status of children. Cavitatory pulmonary tuberculosis was conventionally the disease of older children and so also bone tuberculosis. However with sensitization due to early exposure to mycobacteria, it may also occur in early childhood and we have seen it occasionally even in infancy. Till recently, pulmonary tuberculosis formed more than 85% of childhood tuberculosis but the advent of HIV infection has resulted in a rise of extra-pulmonary TB, which may constitute about 40% of the total burden.
In TBM, focal lesions are often hidden as long as primary disease is under control and new signs appearing thereafter may suggest that these occult lesions have now become overt. Subacute or chronic encephalopathy may manifest after withdrawal of steroids in the treatment of acute meningitis. The widespread use of BCG vaccine may have influenced the manifestations of tuberculosis in vaccinated children. Neurotuberculosis is three times more common in non-vaccinated than vaccinated children.[11] The manifestations of neurotuberculosis may be atypical and less specific in vaccinated children.
Diagnostic difficulties in childhood tuberculosis
A combination of small number of bacilli and inaccessible sites for bacteriological confirmation makes diagnosis of childhood tuberculosis rather difficult. Circumstantial evidence is often the basis of probable diagnosis. Newer tests are available but with significant limitations.
Interpretation of the tuberculin test
BCG vaccination status can be ignored while interpreting tuberculin test. More than 50% of children demonstrate negative tuberculin test after BCG vaccination. From amongst those who are positive initially, majority become negative within a year of vaccination while the remaining do so within the next 2-3 years.
Ideally 1 TU of PPD with tween 80 as a preservative should be used for testing. 5 TU without preservative may be acceptable. Use of higher strength PPD (10 TU) should be avoided, as the reaction is dose dependent. There exists a large inter-observer difference in reading the test to the tune of 15-50%. It is the induration and not just the erythema that has to be measured. Soft induration with slight edema and painless pink erythema suggests reaction to BCG vaccine while hard induration with red or purplish painful erythema denotes Koch's phenomenon. There may be formation of bullae or ulceration.
A cut off point of 10 mm is considered a result of natural infection on the basis of epidemiological studies. However tuberculin reactions follow a bell shaped curve with few children showing higher reactions even to artificial infection (BCG vaccine) and some reacting less than 10 mm to natural infection. In an immunocompromised host, a reaction of 5 mm is considered suggestive of natural infection. Initial reaction of < 10mm and increasing by 6 mm to > 10 mm within two years is considered evidence of recent infection.[12]
Bacteriology
It is the gold standard for diagnosis of tuberculosis. However, the paucibacillary nature of childhood disease and difficulty in obtaining a proper sample for testing are impediments to diagnosis. Nevertheless every attempt must be made to confirm the diagnosis by the available bacteriological tests.
It is possible to detect bacilli on smear in specimens obtained through fine needle aspiration of lymph nodes, besides the additional cytological information. Sputum may be available in older children especially in those with chronic cavitatory lesion or bronchiectesis. ZN stain is positive only if the number of bacteria is more than 10, 000 per ml of specimen. Unfortunately the yield is less than 20%. Though auromine staining increases the yield, it needs a fluorescent microscope.
Gastric aspirates are used in lieu of sputum in younger children. Early morning samples should be obtained. Initially the stomach contents are aspirated and then small amount of sterile water are injected into the stomach and then re aspirated and added to the specimen. Since gastric acidity is poorly tolerated by tubercle bacilli neutralization of the specimen should be performed immediately with 10% sodium bicarbonate or 40% anhydrous sodium phosphate. With careful attention to details and meticulous technique, tubercle bacilli can be detected in 70% of infants and 30% of older children. Broncho-alveolar lavage (BAL) is more invasive and its yield is poorer than an ideally collected gastric aspirate.[13] An Indian study reported no difference in isolation rates from gastric lavage and BAL.[14]
Conventional culture method uses LJ medium. It is less efficient and labor intensive. The yield is about 30-50%. BACTEC is a radiometric assay that is quick but not superior to a conventional culture method. Septi-check AFB system is more sensitive than other methods but to date no method has proved to be significantly superior to conventional culture. Mycobacterial growth indicator tuber system (MGIT) is a promising test still under research.
Bacterial detection in CSF and other serous fluids is less than 10%. Biochemical tests are used to differentiate TB meningitis from other forms of intracranial infections. CSF ADA values of >8 U/L has 92% specificity and 68% sensitivity for diagnosis of TBM. It has however not proved to be a satisfactory method of distinguishing different neurological infections in infancy and childhood.
Molecular diagnosis
PCR and other amplification techniques were initially thought to be very promising tools for the rapid diagnosis of tuberculosis; however several caveats remain. Contamination of samples by products of previous amplification and presence of inhibitors in the sample may lead to false positive or false negative results. PCR sensitivity show wide ranges of sensitivity between 4-80% and specificity of 80-100% and hence such tests require careful clinical appraisal and judgment.[15]
Serodiagnosis of TB
The challenge is to find an antigen that will be highly sensitive and specific. Since mycobacterial cell wall is made up of numerous proteins, lipids and carbohydrates, there are a number of antigens and epitopes to which the immune system may respond. In addition there may be sensitization due to cross-reacting antigens of environmental mycobacteria or BCG. Antibodies to 38 kDa antigen are elicited during the advanced stage of illness and to 14 kDa and 19 kDa in the early stages. Twenty percent of smear positive and large proportion of smear negative and extrapulmonary TB are seronegative and hence serology cannot be recommended. A60 IgG was positive in only 30% and 38 kDa antibody in only 50% of culture positive cases.[16]
Management issues
Poor patient compliance to anti-tuberculosis therapy is the major contributory factor to failure of the tuberculosis control program and has led increasingly to drug resistance.[17] As a way of achieving high rate of completion of treatment and preventing drug resistance, direct observed treatment has been unanimously recommended in the treatment of tuberculosis. Supervised administration of rifampicin-containing short- course chemotherapy for 6 months (DOTS) has been shown to be successful in as many as 78% of bacteriological confirmed cases compared to only a 38% success rate in non-supervised therapy. Targets set by WHO of 78% detection rate and 85% cure rate is likely to be achieved only by 2013 rather than the projected 2005, as DOTS coverage has increased at a slower rate than anticipated. HIV-infected patients with TB also respond well to short-course therapy; relapse rates are low and similar those of HIV-negative patients.
DOTS is replacing inferior treatment protocols but still is being used in fewer than 40% of estimated new TB cases. Misconceptions threaten to undermine continued success in tuberculosis control. The first misconception is that treatment observation is unnecessary. Treatment observation needs to be made more patient-friendly, but must not be abandoned. The second misconception is that health care reform will strengthen tuberculosis control. TB control is essentially a management problem. Greater accountability of governments, donors and providers is essential. A third misconception is to focus on treating multi-drug-resistant tuberculosis (MDRTB) cases without addressing the root causes of MDRTB. While it is important clinically and epidemiologically in some contexts to care optimally for patients with MDRTB, it is more important to address the causes of MDRTB and to fix these. The fourth misconception is an inordinate concern for sustainability. Delaying assistance will make implementation and sustainability in the future more difficult. Tuberculosis control is remarkably inexpensive and cost-effective, but efforts will fail unless programs have the ability to hire staff, purchase supplies, and contract for services efficiently. Critical issues for the future of tuberculosis control are sustained funding, technical rigor, and good management.[18]
In most of the areas, DOTS is a clinic based service and hence the cost to patients in developing countries is high. Several alternative models have been tried. Trained community lay supervisors have proved successful in South Africa. Using a patient nominated family member, as the drug supervisor is more convenient for patients. Other beneficial strategies include reminder cards sent to defaulters, monetary incentives offered to patients and increased supervision of TB clinic staff. It is not possible to determine from current trials whether health education by itself leads to better adherence to treatment. Even though DOTS is widely advocated as the most cost-effective means of ensuring completion of therapy, no completed trials could be found which confirm or refute this view.[19]
Problems of MDRTB
The early diagnosis of MDRTB and case management by experienced teams from the outset remains the best hope for these patients.[20] Adequately supervised and prolonged combination chemotherapy is essential, with drug choice governed mainly by quality-controlled in vitro drug susceptibility data. There is a more limited role for surgery and immunomodulating therapy, like gamma-interferon, may be a useful adjunct. Clearly the most important therapeutic modality for MDRTB is to treat drug-sensitive TB correctly in the first instance and prevent the emergence of resistant TB. Indian studies also consider DOTS as an effective measure of preventing MDRTB.[21]
For populations in which MDRTB is endemic, the outcome of the standard short-course chemotherapy regimen remains uncertain. Unacceptable failure rates have been reported and resistance to additional agents may be induced. As a consequence there have been calls for well-functioning DOTS programmes to provide additional services in areas with high rates of multidrug-resistant tuberculosis. These "DOTS-plus for MDRTB programmes" may need to modify all five elements of the DOTS strategy: the treatment may need to be individualized rather than standardized; laboratory services may need to provide facilities for on-site culture and antibiotic susceptibility testing; reliable supplies of a wide range of expensive second-line agents would have to be supplied; operational studies would be required to determine the indications for and format of the expanded programmes; financial and technical support from international organizations and western governments would be needed in addition to that obtained from local governments.[22]
Potential Approaches to Immunotherapy
In MDRTB, immunotherapy may be an adjuvant to chemotherapy. A vaccine that stimulates selectively a Th 1 response (M. vaccae) has been tried with some success.[23]
Cytokines that enhance bacillary elimination such as IL-2, IL-12 or gamma-IFN are considered for use though high cost, toxicity and large volume required for injection are impediments for its implementation. Neutralizing antibodies to immunosuppressive cytokines such IL-4, IL-10 or to cytokines that lead to immunopathology such as TNF-alpha are possible interventions.
Global emergency of tuberculosis
In view of the global emergency of tuberculosis, WHO "Stop TB" campaign has called for universal adoption of its directly observed therapy, short course strategy (DOTS). Also, through the Massive Effort Against Diseases of Poverty, several international agencies are urging the establishment of effective control programmes worldwide.[24]
References
1. WHO Global tuberculosis control. WHO Report 2001, Geneva, Switzerland: World Health Organization; 2001.
2. Kalaudt K. TB a global emergency WHO report on the TB epidemic, Gemeva, WHO 1994; 3.
3. Donald PR. Children and tuberculosis: protecting the next generation Lancet 1999; 353: 1001.
4. Donald PR. Childhood tuberculosis: out of control Curr Opin Pulm Med 2002; 8: 178-82.
5. WHO Report HTM/TB/2004: 331.
6. Bandera A, Gori A, Catozzi L, Degli Esposti A, Marchetti G, Molteni C et al. Molecular epidemiology study of exogenous reinfection in an area with a low incidence of tuberculosis. J Clin Microbiol 2001; 39 : 2213-2218.
7. Van Rie A, Warren R, Richardson M, Van Der Spuy G, Victor T et al. Exogenous reinfection as a cause of recurrent tuberculosis after curative treatment. N Engl J Med 1999; 341 : 1174-1179.
8. Flynn JL, Chan J. Immunology of tuberculosis. Annu Rev Immunol 2001; 19: 93-129.
9. Ellner JJ. Suppressor adherent cells in human tuberculosis. J Immunol 1978; 121 : 2573-2578.
10. Bentwich Z, Kalinkovich A, Weisman Z, Borkow G, Beyers N, Beyers AD. Can eradication of worms change the face of AIDS and tuberculosis Immunol Today 1999; 20 : 485-487.
11. Somu N, Vijaysekharan D, Balchandran A, Subramanyan L, Chandrabhushanam A. Tuberculous disease in a pediatric referral center - 16 years experience. Indian Pediatr 1994; 31: 1245-1249.
12. American Thoracic Society. The tuberculin skin test. Official ATS statement. Am Rev Respir 1984; 44.9
13. Abadco DL, Steiner P. Gastric lavage is better than bronchoalveolar lavage for isolation of M.TB in childhood pulmonary disease. Pediatr Infect Dis J 1992; 11 : 735-738.
14. Meenu Singh, NV. Ahsan Moosa, Lata Kumar, Meera Sharma Gastric lavage and broncho-alveolar lavage in the bacteriological diagnosis of childhood pulmonary tuberculosis. Indian Pediatr 2000; 37 : 947-951.
15. Smith KC, Starke JR. Eisencah Detection of MTB in clinical specimens from children using PCR. Pediatrics 1996; 97 : 155-160.
16. Swaminathan S, Umadevi P, Shantha S, Radhakrishnan A, Datta M Serodiagnosis of TB in children using two ELISA kits Indian J Pediatr 1999; 66 : 837-842.
17. Meenu Singh Adherence to anti-tuberculosis treatment. Indian Pediatr 1999; 36 : 1285-1286.
18. Frieden TR, Driver CR. Tuberculosis control: past 10 years and future progress. Tuberculosis (Edinb) 2003; 83(1-3) : 82-85.
19. T. Jacob John. Tuberculosis control without protection from BCG vaccine. Indian Pediatr 2000; 37 : 9-18
20. Drobniewski FA, Balabanova YM. The diagnosis and management of multiple-drug-resistant-tuberculosis at the beginning of the new millenium. Int J Infect Dis 2002; 6: S21-31.
21. Prabhakar R. Tuberculosis control in India-past, present and future. J Indian Med Assoc 2000; 98(3): 123-125.
22. Bastian I, Rigouts L, Van Deun A, Portaels F. Directly observed treatment, short-course strategy and multidrug-resistant tuberculosis: are any modifications required Bull WHO 2000; 78(2) : 38051.
23. Stanford JL, Stanford CA, Grange JM, Lan NN, Etemadi A. Does immunotherapy with heat-killed Mycobacterium vaccae offer hope for the treatment of multi-drug-resistant pulmonary tuberculosis Respir Med 2001; 95(6) : 444-447.
24. Grange JM, Zumla A. The global emergency of tuberculosis. JR Soc Health 2002; 122(2): 78-81.(Amdekar YK)