TB transmission directly relates to the intensity of exposure8–14 of the child to the source of TB infection, the clinical severity and infectiousness of disease in the index case, nutritional status, presence of a Bacillus Calmette-Guérin (BCG) scar, physical long term proximity to the individual with TB, household size and duration of cough in the index case as well as magnitude of TST response. Determining which children with LTBI are at the greatest risk of progressing to active TB would allow the identification and treatment of at-risk individuals and reduce the number of active TB cases. This would be a major step forward for TB control programs. Not every child exposed to a contact becomes infected and the probability depends on several factors namely: contagiousness of the source, bacteriological status smears positive, virulence of the bacilli, environmental risks such as: infection control measures, air circulation, exposure to sunlight, proximity of the infected person.

Recently, the possibility that a short occasional contact with an infectious adult (15–20min) is sufficient to generate TB infection and disease has been demonstrated.15

Infected children represent a large proportion of the pool from which TB cases will arise, knowledge of the factors that influence TB infection in children allows intervention in community tuberculosis transmission and adapting TB control activities. One of the global indicators is the proportion of children, among those eligible, under 5 years of age and household contacts of TB cases, started on LTBI treatment.

To avoid the administration of LTBI treatment to those without the infection,12–14 thus potentially increasing costs and adverse events, it is crucial to improve tools to maximize the predictive value of existing tests. This is of particular importance in high incidence HIV co-infection countries because both currently available tests, TST and IGRA, have reduced sensitivity in immune compromised patients so have a low predictive value for progression to active TB. For these reasons, LTBI screening should be reserved for those children who are at a sufficiently high risk of progressing to disease. Such high-risk children may be identifiable using multivariable risk prediction models, which incorporate test results with traditional risk factors, and using serial testing to resolve underlying phenotypes.

A latent infection is defined by a positive score to an immune-based assay, either the tuberculin skin test (TST) or the interferon-γ release assays (IGRA), which are based on the recognition of mycobacterial antigens16 in a healthy person without lesions of active TB in the chest radiological images. TST is quantified by measuring the transverse diameter of skin induration resulting from intradermal injection of purified protein derivative, a crude mixture of antigens, many of which are shared by Mt and other Mycobacteria, in particular Bacillus Calmette–Guérin (BCG). IGRA, including QuantiFERON TB Plus (QFT-Plus; Qiagen, Hilden, Germany) and T-SPOT.TB (Oxford Immunotec, Abingdon, UK), are blood assays measuring in vitro IFN-γ production to specific antigen of Mt, not shared with BCG.16–21 Based on the characteristics of the assays, TST is less specific for latent tuberculosis identification in those that are BCG-vaccinated, compared to IGRA. However, IGRA may be less sensitive, in younger aged children. In spite of increasing evidence of potential utility of IGRA, advisory boards recommend the use of TST in children younger than 5 years of age.7,22–24

Systematic review and meta-analysis report the prevalence of active TB in all contacts to be 3.1% (95% CI 2.2–4.4%, I2=99.4%), 1.2% (95% CI 0.9–1.8%, I2=95.9%) in microbiologically confirmed TB with latent TB infection at 51.5% (95% CI 47.1–55.8%, I2=98.9%). Similar findings are replicated looking only at MDR-TB exposure. Incidence was greatest in the first year after exposure year and remains above background incidence for at least 5 years after exposure to a patient with TB.2,3

Considering that children under 5 years old and people living with HIV contacts of TB patients are a high-risk group for developing TB, particularly within the first year, it is mandatory to screen HIV TB co-infection on those groups. Children under 5 years old diagnosed with LTBI need to get preventive treatment or 2 years of radiological follow-up, depending on their probability of developing TB, contra-indications for preventive treatment and willingness to start preventive treatment for 4 months of rifampicin (RIF) only or 3 months of rifampicin/isoniazid. The appropriate regimen is based on the following: Drug-susceptibility results of the presumed source case (if known), coexisting medical illness and potential for drug-drug interactions. In all cases, children need to go on directly observed therapy (DOT) for LTBI Treatment.

Isoniazid (INH) Regimen – There are 2 options9–11 for treatment with INH: 9 or 6 month regimen. The 9-month regimen is preferred because it is more effective. Treatment for LTBI for 6 months rather than 9 months may be more cost-effective and result in greater adherence by patients; therefore, health care providers may prefer to implement the 6-month regimen rather than the 9-month regimen. Every effort should be made to ensure that patients adhere to LTBI treatment for at least 6 months. The preferred regimen for children aged 2–11 years is 9 months of daily INH.

12-Dose (Isoniazid and Rifapentine [RPT]) Regimen – The directly observed 12-dose once-weekly regimen of INH and RPT is recommended as an option equal to the standard INH 9-month daily regimen for treating LTBI in otherwise healthy people, 12 years of age and older, who were recently in contact with infectious TB, or who had TST or IGRA test or blood test for TB infection conversions, or those with radiologic findings consistent with healed pulmonary TB. The 12-dose regimen can be considered for other groups on a case by case basis when it offers practical advantages, such as completion within a limited timeframe. The regimen may be used in otherwise healthy HIV-infected persons, 12 years of age and older, who are not on antiretroviral medications. It may also be considered for children aged 2–11 years if completion of 9 months of INH is unlikely and hazard of TB disease is great. The 12-dose regimen is NOT recommended for the following individuals: (1) children younger than 2 years of age; (2) children HIV/AIDS who are taking antiretroviral therapy (ART); (3) people presumed to be infected with INH or rifampin-resistant M. tuberculosis.

A 4-month regimen of RIF can be considered for children who cannot tolerate INH or who have been exposed to INH-resistant TB. It should not be used to treat HIV-infected persons taking some combinations of ART.

The choice between the 12-dose regimen and other recommended LTBI treatment regimens depends on several factors, including: feasibility of DOT, resources for drug procurement and children monitoring, considerations of medical and social circumstances that could affect patient adherence, preferences of the patient and prescribing health care provider.

Childhood tuberculosis is characterized by a nonspecific symptomatology. Therefore it is very important to know the history of contacts of the child with pulmonary or laryngeal TB patients that are usually close family members. In fact a child with active TB represents a sentinel event, typically reflecting ongoing transmission in the community.

M. tuberculosis may infect nearly any organ; however, the most common clinical manifestations of disease are found within the thoracic cavity and peripheral lymph nodes. The intrathoracic manifestations are mediastinal or hilar lymphadenopathy and pulmonary parenchymal lesions. More rarely, the pleura or pericardium are additional sites. Isolated intrathoracic lymphadenopathy may be detected early after infection and is often not associated with symptoms. Later, this inflammatory process increases, leading to an enlargement of lymph nodes with complications characterized by an obstruction or compression of the small airways which may be associated with cough, wheezing, or dyspnea; the caseous necrosis of the lymph nodes that may erupt into the airway, leading to bronchopneumonia and manifesting with cough, dyspnea, malaise, and fever; hypersensitivity reactions may also occur, including pleural effusions, which may provoke symptoms of chest pain, fever, and weakness. This nonspecific symptomatology can easily be confused with bacterial or viral causes of pneumonia. Chronic cough may apply, but young children may rapidly progress to disease after exposure, especially if immune depressed and/or very young.41,42

Extrathoracic manifestations make up approximately 20–40% of TB cases, although concomitant overlap with pulmonary disease can occur. Extra-pulmonary symptoms depend on the TB localization and therefore they may greatly vary. M. tuberculosis can reach the central nervous system causing meningitis and tuberculoma; the eye causing uveitis and phlyctenular conjunctivitisa; the ear causing optic or nasopharyngeal chronic suppurative otitis media, mastoiditis, tonsillitis, laryngitis; the heart causing cardiac pericardial effusion; the gastrointestinal apparatus causing abdominal peritonitis, enteritis, involvement of lymph nodes, visceral involvement; the genitourinary involvement causing interstitial nephritis and glomerulonephritis; the bone causing osteoarticular and vertebral osteomyelitis; the lymphatic adenopathy (cervical localization is the most common); the skin causing different manifestations such as chancres, warts, lupus vulgaris like lesion, pustulonodular lesions, erythema induratum. The most commonly involved extrathoracic sites are the peripheral lymph nodes or the central nervous system. Lymphadenitis often manifests in the cervical regions with enlarged, painless lymph nodes. A solitary elastic node without erythema or warmth is a common manifestation of the disease. Contiguous nodes may enlarge and become palpable and the lesion then grows tangled and immovable; fistulas may also form. The most worrying complication of TB is when it is localized in the central nervous system because it leads to a significant morbidity or mortality occurring in approximately 50% of cases.43,44 The start is subacute and the symptoms nonspecific during the early stages, such as irritability, fever, and anorexia, and possible focal respiratory or gastrointestinal symptoms. As disease progresses, findings of meningitis become apparent, including vomiting, altered consciousness, convulsions, meningismus, cranial nerve palsies, or signs of raised intracranial pressure. It is crucial to obtain an early diagnosis which is the only tool to improve the clinical outcomes.

Imaging studies play an important role in the diagnosis of intrathoracic TB. Chest radiographs, although commonly used, are relatively nonspecific per TB diagnosis.45,46 Indicative findings consist of: thoracic lymphadenopathy; air-space disease, which may be indistinguishable from other causes of pneumonia; miliary nodules or cavitation. Computer-aided detection for pulmonary TB has not yet been validated among children.

Depending on the extra-pulmonary TB localization, ultrasonography, computer-aided detection, magnetic resonance can be used.

The diagnosis of TB in children is challenging due to the paucibacillary nature of disease and the subtle or nonspecific radiographic findings.47 Up to the present time, an accurate diagnostic test for pediatric TB is not available. Therefore, it is crucial for clinicians to suspect of the possibility of active TB and to look for all the clinical symptoms that may be the basis for a diagnosis. Indeed, given the poor sensitivity of the current diagnostic tools, TB diagnosis is often based on clinical means.

TB diagnosis can be “confirmatory” if based on the direct detection of the pathogen; alternative approaches include the evaluation of the histopathological lesions or the detection of the host immune response to the pathogen.

Direct pathogen-based tests include Mt smear microscopy, culture, nucleic acid amplification tests (NAATs). The mycobacterial culture is the gold-standard test for TB. The sensitivity is low (7–40%) due to the paucibacillary nature of disease.48,49 It takes up to 6 weeks for positive growth. When Mt is isolated, it is important to perform drug-sensitivity testing against first-line TB drugs to start with, and against second-line TB drugs if needed. The smear microscopy has a lower sensitivity compared to culture and therefore its accuracy is limited in pediatric TB cases. The newer GeneXpert MTB/RIF (Cepheid, Sunnyvale, CA, USA) assay is a NAAT that detects M tuberculosis DNA and concomitant resistance to rifampicin.50 The GeneXpert MTB RIF assay requires less training and infrastructure to perform and interpret than other commercially available molecular assays, produces results in about 2h which has contributed to its wide spread implementation throughout high-TB burden settings.51 GeneXpert only has a pooled sensitivity of approximately 66% compared with TB culture in pediatric populations. Repeated sampling offers incremental yield.52–54

For pulmonary TB, the collection of sputum may be very difficult, especially for children younger than 7 years. Semi-invasive techniques, such as gastric aspiration and lavage, or sputum induction with or without nasopharyngeal aspiration, may be required. The use of the string test in which a gelatin capsule containing a nylon string is swallowed and later retrieved for TB culture49 is an alternative method of obtaining respiratory specimens.55

In extrapulmonary TB, site-specific specimens for TB culture are often collected, such as cerebrospinal fluid, lymph node aspirates, and other tissue specimens. Mycobacterial blood cultures have a lower accuracy in children compared with adults.49,56 Histopathological diagnosis is used, but the accuracy is not well characterized.

Improved diagnostic tools are urgently needed for children.57 New approaches have been evaluated in pediatric studies using nonrespiratory specimens, such as blood-based assays, such as the T-cell activation marker which use host immune responses as a diagnostic biomarker58 and transcriptomic studies, which hold promise as a diagnostic and prognostic marker.18,59–61 Other biomarkers that have shown promise in adult populations, such as the antibodies in lymphocyte supernatant assay and the urinary lipoarabinomannan assay or IP-10, have not shown consistent results in pediatric populations.62–67 Urine is an interesting diagnostic specimen, especially for children because easy to obtain and safe to handle for health care personnel. These characteristics would provide an extraordinary advantage as a point-of-care diagnostic test, and further work in this area is needed.