Epidemiology and clinical management of tuberculosis in children in Canada

Abstract

Although often regarded as a foreign disease, latent tuberculosis or tuberculosis disease will be encountered in many clinical situations by the Canadian child health practitioner. There are key differences between tuberculosis in children and adults. In the present article, the changing epidemiology of tuberculosis in children in Canada and around the world, the pathogenesis of infection, diagnostic tests, and clinical management of childhood latent tuberculosis and tuberculosis disease are reviewed.

There are many situations in which a Canadian-based paediatrician must manage latent tuberculosis (TB) or TB disease. In the present review, we discuss the epidemiology, diagnosis and management of childhood TB in Canada. It is, in part, derived from and supplements the Canadian Tuberculosis Standards, 7th Edition, which is available open-access online (1).

GLOBAL EPIDEMIOLOGY OF TB AMONG CHILDREN

The WHO estimates that, worldwide, there were 9.0 million new TB infections and 1.5 million deaths in 2013, making TB the second-leading infectious cause of death (after HIV/AIDS) (2). An estimated 550,000 children <15 years of age developed TB disease in 2013, and 80,000 non-HIV infected children died from the disease (2). Child TB cases are under-reported, and the true global burden is likely to be even higher than the WHO estimate (3). Nearly 95% of TB cases in the world occur in low-income countries, and 80% of the disease burden is in 22 countries, mainly in Asia and sub-Saharan Africa (2).

EPIDEMIOLOGY OF TB IN CANADIAN CHILDREN

From 2000 to 2012, the number of cases of TB disease in Canada decreased from 1724 to 1686, and the overall incidence is currently approximately 4.8 cases per 100,000 population compared with the global incidence of 122 per 100,000 (1). From 2005 to 2012, an average of 95 cases of TB were reported annually in children <15 years of age: the highest paediatric rate was in infants <1 year of age (1). In Canada, paediatric TB is largely a disease affecting foreign-born children, children of foreign-born parents, and First Nations and Inuit children (1,4). Based on the most recent available data, Nunavut has rates of approximately 85 per 100,000 population <15 years of age (5,6).

PATHOGENESIS

Infection typically occurs when a child inhales Mycobacteria tuberculosis exhaled by an individual with pulmonary or laryngeal TB. Primary infection due to skin inoculation, ingestion into the gastrointestinal tract or congenital infection due to transplacental transmission occurs very rarely (7,8).

After inhalation, a primary reaction often occurs consisting of a small parenchymal lesion in the lung with regional lymphadenopathy, often perihilar or mediastinal (9). Bacilli may then spread throughout the lungs and pleura, which can result in intrathoracic TB. In addition, bacilli may also travel to the meninges, bone, liver and spleen. Both pulmonary TB and extrapulmonary sites of infection may result in latent TB infection (LTBI), which can reactivate months or years later. Rarely, disseminated disease develops within three to six months from the time of primary infection, especially in children <4 years of age (7).

LTBI AND TB DISEASE

A child with LTBI has a small number of sequestered bacilli in the body that are controlled but not killed by the immune response. Clinically, LTBI is practically defined as a positive tuberculin skin test (TST) or interferon-gamma release assay (IGRA), no findings of TB disease on physical examination, and a chest x-ray that is either normal or only has calcifications. While approximately 10% of untreated adults with LTBI will progress to develop reactivated disease over their lifetime, up to 40% of children <12 months of age reactivate and present with TB disease, much of which is extrapulmonary or disseminated (10,11).

TB disease is diagnosed when the clinical signs and symptoms or radiological findings caused by M tuberculosis occur. Although clinicians make a distinction between LTBI and TB disease, these entities exist on a continuum related to bacterial load. Risk factors for progression to TB disease include young age, congenital and acquired immune deficiencies, renal failure, diabetes mellitus and receipt of immune suppressing medications including biologic anti-tumour necrosis factor alpha agents (1,8,12–15).

CLINICAL PRESENTATION OF TB DISEASE

There are two main ways in which children in Canada are diagnosed with TB disease (16). Frequently, an asymptomatic child is found to have an abnormal chest x-ray when evaluated following contact with an infectious adult or adolescent (17). Alternatively, children may also present with symptoms and signs involving any organ system. The presentation of symptomatic TB disease varies according to age: adolescents and older children typically experience fever, weight loss and night sweats, and those with pulmonary disease may exhibit a productive cough, hemoptysis and chills (18). In contrast, younger infants often have nonspecific symptoms and signs, with miliary disease and TB meningitis being more common than in older children (10). TB meningitis is most likely to occur in children <5 years of age, especially infants <12 months of age. In contrast to bacterial meningitis, TB meningitis has a more subacute presentation over days or weeks that includes fever, behavioural changes and headache. Gastrointestinal TB may mimic inflammatory bowel disease. TB may affect any bone in the body and symptoms mimic a subacute or chronic bacterial osteomyelitis, with spine involvement in nearly 50% of paediatric skeletal TB cases (19,20). Children coinfected with HIV and TB experience an accelerated progression from infection to disease (1). While coinfected adults often have atypical presentations with extrapulmonary disease, children with HIV and TB usually present similarly to HIV-negative children of the same age group. However, their TST is often negative (21).

DIAGNOSTIC TESTS

TST

In the TST, 0.1 mL of purified protein derivative (PPD) is injected intradermally into the volar surface of the forearm to evaluate cell-mediated immunity against TB. In an individual who is sensitized from previous exposure to mycobacteria, a reaction occurs at the site of injection, which should be measured 48 h to 72 h after placement and the result recorded in millimetres of induration. PPD has approximately 200 antigens, some of which are common to M tuberculosis, Mycobacterium bovis and other nontuberculous mycobacteria (NTM). Appropriate technique is essential for accurate results; step-by-step guidelines are available in the Canadian Tuberculosis Standards (1). Current Canadian guidelines for interpretation of the TST are presented in Table 1. It can take up to eight weeks after exposure for a positive TST to occur (22). False-negative tests can occur in up to 10% to 20% of immunocompetent children with proven TB disease and can also be caused by the use of corticosteroids, recent viral infections, recent live viral vaccines, skin anergy, young age (<6 months of age) and incorrect administration of PPD (23).

TABLE 1.

Canadian guidelines for interpreting a tuberculin skin test (TST)

Factor	Guideline
Time	Results should be read by a trained health professional 48 h to 72 h after administration Size of induration (palpable, raised, hardened area) Positive predictive value Risk of developing disease if person is truly infected
Factors used to interpret a positive TST
Size of induration
	TST result	Situation in which reaction is considered to be positive
	0–4 mm	Generally, considered negative and no treatment is indicated Child <5 years of age and at high risk for TB infection (refer to Figure 1)
	≥5 mm	HIV infection Contact with infectious TB case within the past two years Presence of fibronodular disease on chest x-ray (healed TB and not previously treated) Organ transplantation (related to immune suppressant therapy) Tumour necrosis factor-alpha inhibitors Other immunosuppressive drugs such as corticosteroids (eTuivalent of ≥15 mg day of prednisone for ≥1 month; risk of TB disease increases with higher dose and longer duration) End stage renal disease
	≥10 mm	Criterion of 10 mm has a sensitivity of 90% and specificity of >95% – recommended criterion for most clinical situations in Canada All others, including the following specific situations: ○ TST conversions (within two years) ○ Diabetes, malnutrition (<90% ideal body weight), cigarette smoking daily alcohol consumption (>3 drinks day) ○ Silicosis ○ Hematological malignancies (leukemia, lymphoma) and certain carcinomas (eg, head and neck)

Table adapted from reference 1. Positive predictive value: probability that a positive test result represents the true presence of tuberculosis (TB) infection; Risk of developing active TB: Many medical conditions and treatment regimens can increase risk of reactivation (HIV, diabetes, renal failure, malnutrition, certain cancers, alcohol overuse and cigarette smoking)

Sensitization to PPD may occur from previous infection with NTM or previous receipt of the Bacille Calmette-Guerin (BCG) vaccine. Fewer than one-half of all infants who receive the BCG vaccine at birth will have a positive TST between nine and 12 months of age, and the majority will be nonreactive by five years of age (24). BCG vaccination after 12 months of age has been associated with prolonged persistent false-positive TSTs (25). However, because BCG is most commonly administered in areas with high rates of endemic TB, a history of BCG is generally ignored when administering and interpreting a TST.

Interferon-gamma release assays

IGRAs measure in vitro interferon-gamma production by circulating T lymphocytes in response to M tuberculosis antigens that are absent from M bovis, BCG and most NTM. Commercially available IGRAs include the QuantiFERON-TB Gold test (QFT-G) (Cellestis Ltd, Australia) and the T-SPOT TB assay (T-SPOT; Oxford Immunotec Ltd, United Kingdom). For culture-confirmed disease, the sensitivity of IGRA testing in adults is similar to the TST and shows a higher degree of specificity (26–33). There are few data for children <5 years of age and the tests should be used with caution in this age group. Children in this age group manifest a lack of T cell response from an immature immune system, which likely contributes to false negatives in IGRA testing with a similar mechanism as TSTs (34). In a Canadian paediatric study, the QFT-G used in addition to the TST did not detect a significant number of extra cases of LTBI in high-risk close contacts, and the agreement between the tests was good (35). However, many lower-risk contacts and those without contact history were TST positive and QFT negative – a finding that was much more likely if the patients had received the BCG vaccine (35). The use of an IGRA instead of a TST or as a ‘rule-out’ test in TST-positive individuals in these lower-risk groups would likely prevent unnecessary treatment of patients who have a false-positive TST rather than LTBI (36).

Current Canadian recommendations state that both the TST and IGRA are acceptable for diagnosis of LTBI (1). However, IGRA testing is preferred over TST when a person has received BCG vaccine after 12 months of age and/or has received BCG more than once, and in situations in which there is reason to believe an individual will not return for TST reading. In children with suspected TB disease, TSTs and IGRAs may be used in combination with imaging and sample collection for microbiology to assist in making a diagnosis of TB. Neither a negative TST nor negative IGRA test rules out TB disease.

Sample collection and microbiology

In suspected TB disease, it is important to obtain samples for diagnosis and drug sensitivity testing. Sputa should be obtained if the child can expectorate; however, young children typically <6 to 10 years of age often cannot produce adequate samples and, thus, morning gastric aspirates are often collected as an alternative to collect pulmonary material that was swallowed throughout the night. Samples are collected on three consecutive early mornings in hospital or on an outpatient basis (37). The procedure is described in the Canadian TB Standards (1). Acid-fast staining of the aspirate sample is typically negative and the sensitivity of culture has shown considerable variation, ranging from 17% to 50% for three samples when clinical signs of TB disease are present (1,18,38).

Induced sputa are obtained in older children by deep expectoration after the administration of hypertonic saline. For younger children, nebulization of hypertonic saline is followed by a timed nasopharyngeal aspirate (1).

When TB is suspected outside the lung parenchyma, samples should be obtained from the suspected tissue via biopsy of lymph nodes, pleura or other organs, lumbar puncture and urine collection.

Use of liquid medium culture techniques for M tuberculosis is now standard in all TB laboratories in Canada; the average time-to-growth detection is now 10 to 14 days. Nucleic acid amplification tests (NAAT) use primers specific to TB and have excellent sensitivity for differentiating TB from NTM on samples that are acid-fast bacilli (AFB)-positive. However, the sensitivity of NAAT on AFB negative samples is unacceptably low.

The GeneXpert MTB/RIF (“Xpert”) (Cepheid Inc, USA) is a rapid gene amplification test that can be used directly on samples rather than only on positive cultures to simultaneously detect the presence of M tuberculosis and resistance to rifampicin. Results are available within 2 h. The overall pooled sensitivity of Xpert MTB/RIF compared with culture in children was 66% when expectorated sputum or induced sputum was used, and 66% for gastric lavage or aspiration with a specificity of at least 98%. The sensitivity to detect rifampicin resistance in specimens from children was 86% (39). Results of Xpert and other rapid molecular tests for direct sensitivity testing should be confirmed by conventional culture methods (1).

DIAGNOSTIC IMAGING

Because obtaining culture-positive samples in children can be difficult, diagnostic imaging in conjunction with TST and/or IGRA along with clinical correlation is a cornerstone in the diagnosis of TB disease. Both frontal and lateral chest radiographs should be obtained to identify hilar lymphadenopathy and parenchymal consolidation. Parenchymal lesions may occur in any part of the lung, but are most commonly found at the apices. Cavitation is rare in young children, but may occur in older children and adolescents (40). Other diagnostic imaging tests may be used as clinically indicated. An experienced paediatric radiologist should review the images.

DIAGNOSIS OF TB DISEASE: GENERAL CONSIDERATIONS (TABLE 2)

TABLE 2.

Paediatric tuberculosis (TB) disease: essential points

Who is most likely to become infected with TB in Canada? Foreign-born or Aboriginal children Children with foreign-born parents Those who travel to endemic countries, especially with prolonged stay/contact with local population

When to suspect TB disease: Epidemiology: as above. May not always be apparent Symptoms and signs: more common in older children and adolescents ○ B symptoms: fever, weight loss, night sweats ○ Pulmonary: productive cough, hemoptysis, chills ○ Extrapulmonary: nonspecific, can present in any organ system Note: infants and young children commonly have nonspecific signs/symptoms

What tests should we consider/order? Posteroanterior and lateral chest radiograph Other imaging as clinically indicated Obtain TB cultures as indicated: sputum (induced if possible) for AAFB, gastric aspirates, biopsies, cerebrospinal fluid. Ensure sent without formalin; liaise with TB laboratory TST/IGRA: may be negative in 10%; consider it to be supportive evidence if positive

How do we begin management of TB disease? Ensure appropriate isolation Notify and involve local public health unit Involve TB team and experts See Canadian TB Standards for details (1)

AAFB Alcohol and acid-fast bacilli; IGRA Interferon-gamma release assay; TST Tuberculin skin test

Because childhood TB is paucibacillary, it is often culture negative. Diagnosis is often made on the basis of immunological evidence of tuberculosis exposure (positive TST or IGRA), characteristic chest x-ray abnormalities and epidemiology. Criteria for the diagnosis of confirmed, probable and possible childhood intrathoracic TB have been proposed by an international expert panel in the context of clinical research (41).

TREATMENT OF TB DISEASE

A multidisciplinary approach is the best way to achieve optimal treatment as well as to deal with side effects and questions that arise during therapy. There is evidence that outcomes are better when treating physicians have recent and substantial experience in managing TB (42). However, in settings where this is not possible, management should involve discussion with a TB expert at minimum (Table 2).

Treatment for TB disease should always include multiple drugs to prevent the emergence of resistant strains. Specimens should always be obtained before starting treatment to attempt to grow M tuberculosis. The four commonly used first-line drugs are isoniazid (INH), rifampin/rifampicin (RMP), pyrazinamide (PZA) and ethambutol (EMB); in general, all four should be used initially, pending sensitivities, unless the source strain is known to be drug sensitive. Detailed treatment regimens, including for rare presentations such as TB meningitis, the mode and mechanism of action of each drug, dosages in children and main toxicities, can be found in the Canadian TB Standards (1).

In cases of fully drug-sensitive pulmonary and extrapulmonary TB disease not affecting the central nervous system, patients should receive a minimum of two months of INH, RMP and PZA, followed by four months of INH and RMP. Daily therapy is preferred over intermittent regimens (1). For those who remain sputum culture positive after two months of compliant therapy, treatment should be prolonged to nine months. If sensitivities are not available from either the source case or the patient, the first two months of therapy should also include EMB. Those with extensive or disseminated disease should be treated for longer intervals, often 12 months. Deterioration during therapy is fairly frequent in children and is difficult to differentiate from clinical failure, drug resistance or an alternative diagnosis (43).

Directly observed therapy (DOT) is important in ensuring cure and preventing the emergence of resistant strains and, as a result, may improve survival (42). DOT is recommended for all paediatric patients with TB disease and is mandatory for the treatment of resistant TB (1).

The incidence of multidrug-resistant TB is increasing (44). Risk factors include a history of previous treatment for TB, contact with a known drug-resistant case, origin from or travel to a country with a high prevalence of drug-resistant TB, and failure to respond to a standard regimen. Management should always be directed by a TB specialist and consists of multiple second- and sometimes third-line drugs and longer durations of therapy (1,45–47).

LTBI – SCREENING AND TREATMENT (TABLE 3)

TABLE 3.

Latent tuberculosis infection (LTBI): essential points

Who do I screen for LTBI? Testing should never be routine, always targeted to high-risk groups High risk for infection: ○ Contacts of infectious source cases ○ Adoptees and immigrants from high-burden countries Consider those at high risk of progressing from infection to disease: ○ Pretransplant, HIV infection, pre anti-tumour necrosis factor alpha inhibitors

How do I diagnose LTBI? Positve TST or IGRA. No clinical evidence of disease. Normal chest x-ray

How do I treat LTBI? Isoniazid (INH)-sensitive source case or if source case unknown: INH (approximately 10 mg/kg/day to maximum of 300 mg) for nine months Vitamin B6 if milk- or meat-deficient diet, if breastfed or any concern about diet. Dose: approximately 1 mg/kg up to 25 mg (approximate to one-quarter, one-half or an entire 25 mg tablet) Monitoring: See monthly for assessment for side effects. Instruct patients to discontinue prescription if any anorexia, abdominal pain, vomiting or jaundice. Provide written plan of action for patient if symptoms. Check AST ALT Bilirubin if any symptoms occur If drug resistant source isolate, obtain expert opinion

AST Aspartate aminotransferase; ALT Alanine aminotransferase; IGRA Interferon-gamma release assay; TST Tuberculin skin test

Screening for latent infection should always be targeted: if the TST is used in populations with a true low prevalence of disease, most tests will be false positive. Although IGRAs have greater specificity, false-positive values may occur, especially at the ‘cutoff’ values for the test. Screening should be prioritized in certain circumstances including close-contact cases with infectious adults and children who were born in an endemic country or resided in one for >3 months (1,48). This screening is strongly warranted if there are compounding risk factors such as young age or immunosuppression.

To diagnose LTBI in a child, TB disease should be excluded by clinical history, examination and a chest x-ray. If the strain from the source case is INH sensitive or there is no index case, children with latent disease should be treated for nine months with INH, which reduces the lifetime risk of reactivation by >80% (49). For contacts of INH-resistant rifampin-sensitive LTBI, four months of rifampin are recommended.

INH hepatotoxicity is rare, but does occur, and has led to liver failure and death (50). In addition to monthly assessments for side effects and compliance, patients should be informed to discontinue medication immediately should symptoms (anorexia, vomiting, nausea, jaundice or abdominal pain) arise and should be provided with a clear written plan of action, including contact numbers. If symptoms occur, evaluation should include a physical examination and assessment of liver transaminase and bilirubin levels because children may appear clinically well despite impending liver failure (51).

MANAGEMENT OF CONTACTS (FIGURE 1)

Figure 1).

How to manage a child exposed to infectious tuberculosis. IGRA Interferon-gamma release assay; LTBI Latent tuberculosis infection; TST Tuberculin skin test

All children who are exposed to TB should be assessed for symptoms and have a TST performed. In contact cases in which the risk for developing TB disease is high, including children <5 years of age or immunocompromised children, a clinical evaluation, TST and chest x-ray should be completed. If there is no evidence of disease in high-risk children, they should begin ‘window prophylaxis’ with INH (or rifampin if the source strain is INH-resistant and rifampin sensitive), even if the TST is negative, because these individuals may be at risk for rapid progression to severe disease (10). If a repeat TST remains negative eight weeks after the break in contact with the index case, treatment may be stopped. However, if the second TST is positive (>5 mm induration for a close contact) the child should undergo an additional chest x-ray and be re-evaluated for the presence of disease. If no disease is present, then a full course of treatment for LTBI should be completed.

SUMMARY

Although the incidence of TB has decreased in Canada, both LTBI and TB disease may be encountered, diagnosed and managed by Canadian paediatricians. There are important differences between TB in adults and in children, and it is important for Canadian clinicians to remain familiar with approaches to diagnose and manage TB in this age group and within this changing epidemiological context.