2.1 Presumptive Paediatric TB 

Presumptive Paediatric TB refers to children suspected to be suffering from TB based on any of the following symptoms: persistent fever, cough for more than two weeks, loss of weight. A definite weight loss (>5% loss in the past three months) or failure to gain weight in the past three months despite adequate nutrition with no other apparent cause should prompt detailed history, examination and investigation, including investigations for TB. 

Moreover, presumptive TB cases (Pulmonary or Extrapulmonary) would often have known contact with an infectious TB patient. In a symptomatic child, contact with a person with any form of active TB within the last two years may be deemed significant.

The diagnostic algorithm below gives the recommended pathway for the diagnosis of intrathoracic tuberculosis in children. Experts believe that while the symptoms suggestive of TB can be mimicked by several other diseases; however, if they are properly characterised, the probability of finding those with TB improves. Persistent cough and fever for two weeks or more without a known cause is an excellent clinical marker, particularly if associated with weight loss or a history of exposure to a case of active TB. Weight loss or not gaining weight should always be documented with appropriate and proper weighing. The patient should always be weighed with minimal clothing and without shoes using a tared scale.

All these presumed cases are subjected to further investigation for TB. The initial specific investigation recommended is a frontal Chest skiagram. If a recent good quality chest x-ray film is available, repeat testing is not routinely recommended. This is a significant shift from the earlier guidelines where a sputum smear examination was the initial recommended test.

 

    

Figure. 1 Diagnostic Algorithm for Pulmonary TB in Children

 

2.2. Chest Imaging for Tuberculosis in Children

A plain radiograph with a frontal view of the chest is the initial investigation advised. Radiological findings considered highly suggestive of tuberculosis in the suggested clinical setting include a) miliary shadows – diffuse micronodular shadows affecting both the lungs like a snowstorm. b) intrathoracic lymphadenopathy (usually seen as a dense well-circumscribed ellipsoid or rounded shadows in the hilar or mediastinal regions), c) Chronic fibro-cavitary shadows (usually, but not exclusively, seen in the apical regions).

Findings like consolidations, in-homogenous shadows or bronchopneumonia, etc. are considered non-specific as they may also be seen in other bacterial diseases. The following set of chest skiagrams show the various highly suggestive lesions seen on chest imaging in TB.

 

 

 

 

Figure 2a. Chest Imaging in TB

 

 

Figure 2b. Chest Imaging in TB

 

Chest radiography is also an important tool to support the diagnosis of pulmonary TB in children. In the relevant clinical setting, a diagnosis of TB may need to be made on a clinical basis based on highly suggestive imaging findings of tuberculosis (miliary pattern, hilar or paratracheal lymphadenopathy with or without parenchymal involvement and fibro-cavitary lesions). It is worth remembering that all these imaging patterns also have other differential diagnoses. All presumptive TB cases with these radiological patterns are considered to be probable cases and should be subjected to microbiology to confirm the diagnosis. If microbiological investigations are negative, these children can be deemed to be cases of clinically diagnosed tuberculosis (TB), provided alternative causes are ruled out. Specificity of diagnosis in such situations increases in the presence of pointers like positive TST or history of contact.  

Non-specific findings on chest X-ray refer to patterns other than highly suggestive patterns described above and includes: consolidations, in-homogenous shadows or bronchopneumonia, etc. As these abnormalities can often be seen with other aetiologies like bacterial disease, such patients should be given a course of appropriate antibiotics (Amoxycillin or co-amoxiclav), mainly if the blood counts show neutrophilia.  

Antibiotics like Linezolid or any quinolone should not be used as they have anti-TB action. Clinical clues may prompt earlier investigation for TB in some cases, e.g. widespread bacterial pneumonia is unlikely to be clinically silent (absence of significant distress or fever). In contrast, mycobacterial pneumonia may have disproportionately lesser clinical findings than the extent of radiological changes.

 

 

 

 

 

Figure 2c. Chest Imaging in TB

 

 

 

Suppose the non-specific radiological patterns do not resolve despite adequate antibiotic therapy in a symptomatic child. In that case, they need to be evaluated with Induced Sputum (IS)/ Gastric Aspirate (GA) for M.tb . If negative by NAAT or culture, a further evaluation like CT chest and flexible bronchoscopy for persistent pneumonia are required. The differential diagnosis in such situations is significant, of which TB might be one of the possibilities. These children need to be referred for further evaluation, and a trial of Anti-Tubercular Treatment (ATT) is not justified without ruling out other diseases.

Chest radiographs in a patient who has been treated in the past can have post-TB changes like opacities secondary to pleural thickening, fibro-atelectatic changes or areas of bronchiectasis. These children are usually asymptomatic except in children with post-TB bronchiectasis, where they may show symptoms of suppurative lung disease. Such cases also need expert evaluation to differentiate and confirm disease recurrence if new symptoms come up after successful treatment.   

It is crucial to remember that several factors can lead to both under and overdiagnosis of TB on chest radiology. Leading issues associated with possible errors include technical quality of films like exposure, rotation, motion blur, etc.; inter and intra-individual variations in reading; errors in interpretation due to artefacts; confounders like a thymic shadow, etc.

CT Scan is sometimes necessary but is costly and gives significant radiation. Therefore, it should be used judicially; however, CT is an essential tool for evaluating children with persistent pneumonia or persistent fever without apparent focus. Certain CT patterns that can be highly suggestive of TB, although not diagnostic, are necrotic mediastinal lymphadenopathy, centrilobular nodules with tree-in-bud pattern and cavities with surrounding consolidations. Chest CT scan also may offer an opportunity for CT guided biopsy for tissue diagnosis.

Ultrasonography (USG) of the chest is helpful to assess the pleural fluid collection. It is useful for differentiating thymus and anterior lymph nodes and identifying the best spot for aspiration/ Fine Needle Aspiration Cytology / biopsy in a peripheral lesion.

 

2.3. Microbiological Tests

The new strategy is to test all TB patients upfront for resistance, particularly rifampicin using NTEP-approved rapid NAAT on relevant body specimens. For pulmonary TB, respiratory secretions like self-expectorated sputum or induced sputum or gastric aspirate or lavage, etc., can be used. As the access to samples among children is not easy, there is merit in targeting collection among children after a positive screening on chest radiography. While the patient is immediately subjected to relevant specimen collection and testing if the screening chest imaging shows highly suggestive radiological features, but in case the screening radiograph shows non-specific findings (described before), the respiratory specimen collection for M. tb is advised if the symptom persists even after a full seven days’ course of a potent suitable antibiotic (like Amoxycillin or Co-amoxiclav). Exceptions may need to be made if there is a clinical urgency and the child has a high probability of having tuberculosis.

As most young children cannot expectorate sputum, alternatively, the swallowed sputum is collected from the stomach after a period of fasting (usually 4-6 hrs, preferably overnight) as gastric aspirate. Early morning gastric aspirate is a preferred specimen for most young children with presumptive TB. It needs to be done on an empty stomach and requires skilled staff. While it should ideally be done early in the morning, after overnight fasting (usually feasible in an inpatient), but as an alternative, it can also be collected in ambulatory settings after 4-6 hours of fasting with some compromise on the yield. (See Annexure 2 for Method to collect gastric aspirate; video available at: https://tbcindia.gov.in/index1.php?lang=1&level=2&sublinkid=5305&lid=3415 )

Induced sputum is another sample that can be obtained in children who are unable to produce sputum. Induction of sputum can be easily performed in young children, including infants with an acceptable yield of the sample. It does not require prolonged fasting and should be done with adequate infection control practices as the procedure provokes cough, increasing the risk of transmission. (See Annexure 3 for Method to collect induced sputum; Video available at: https://tbcindia.gov.in/index1.php?lang=1&level=2&sublinkid=5305&lid=3415 ).

A good sputum sample consists of recently discharged material from deep inside the chest (the inner airways) with a minimum amount of oral or nasopharyngeal material, has mucoid or mucopurulent character and should be 2-5 ml in volume. It should be collected in a sterile container after rinsing the oral cavity with clean water. Under the programmatic conditions, a conical tube, like Falcon’s tube, is the preferred container. The collected specimens should be labelled appropriately, safely packaged and transported to the laboratory as soon as possible after collection. If a delay is unavoidable, the samples can be refrigerated to inhibit the growth of unwanted microorganisms and tested no later than seven days after collection. If the patient does not produce any sputum, even after induction, any other available respiratory specimen like bronchoalveolar lavage (BAL), lung aspirate, etc., can be collected by a skilled health care provider depending upon available facilities.

Bronchoscopy and bronchoalveolar lavage are required in select cases of persistent pneumonia for final diagnosis. They may be performed when routine investigations for TB are inconclusive or for children who are drug-resistant TB suspects. Precious or challenging to get specimens like BAL or lung aspirate should also be subjected to culture and drug sensitivity, particularly if the rapid test is negative.

Many children with TB may have a concomitant extra-pulmonary (EP) disease. Specimens from the EP site can also be used to establish a diagnosis. However, no preservative should be used for any extra-pulmonary sample for culture. Moreover, necessary instructions are to be given to the concerned staff for sending the biopsy specimen in normal saline for culture and not in preservatives like formaldehyde as it kills bacilli.

The available initial diagnostic tests are NTEP approved rapid NAAT (Xpert Rif™ - a cartridge-based nested NAAT for detecting M.tb and its resistance to Rifampicin; Truenat™ M.tb  Plus – a chip-based NAAT to detect M.tb and then sequentially test for Rifampicin resistance using M.tb  Rif Dx™). These rapid NAATs can test several different types of clinical specimens like sputum, induced sputum, gastric aspirate or lavage, and various other body fluids. In addition, line probe assays (LPA) are also available under the programme (Hain test/ MTB DRplus™, MTB DRsl™). These tests take up to 72 hours and can be used only on smear-positive sputum specimens or culture isolates for Genotypic DST. This is currently the only WHO, and NTEP recommended rapid test to detect additional drug resistance in MDR-TB and XDR-TB patients. The First Line- LPA thus detects M.tb and the presence of resistance to R and H (MTB DRplus™). A second Line-LPA (MTB DRsl™) is available for testing for resistance to class Fluoroquinolones (FQ) and class second-line injectables (SLI).

A single good quality specimen with the newer rapid molecular tests is usually adequate (unlike two samples needed for the Acid Fast Bacilli smear test). Despite the more recent rapid tests having a much higher sensitivity and specificity than the smear for Acid Fast Bacilli (AFB), the newer tests perform poorly in culture-negative cases. Their sensitivity is less than the culture, which detects between a third to nearly half the cases. Thus, more recent molecular tests are good “Rule–in tests” due to their high specificity. Still, a negative test does not rule out TB disease as approximately half of the paediatric TB cases are culture negative due to the very nature of the disease.

Moreover, studies have shown that newer rapid NAAT tests perform very poorly with pleural or ascitic fluid compared to respiratory specimens, Lymph Node (LN) aspirates, pus and CSF. It is suggested that children with persistent symptoms, non-specific shadows, and negative results of smears and other NAAT samples (GA/IS) should undergo further workup to diagnose persistent pneumonia at appropriate facilities. Many such cases are diagnosed as TB based on the clinical and radiological features after excluding other diagnoses.

As discussed previously, NAAT being a nested PCR, provides additional information about Rifampicin resistance. Furthermore, when Rifampicin resistance (RR) is seen, the patient is offered First Line (FL) and Second Line (SL) Line Probe Assay (LPA). RR detected in a new case with no risk factors for Drug-Resistant Tuberculosis (DR-TB) needs to be retested, only if M.tb detected was very low as that could be a false positive. If there is a discordance in Rifampicin resistance between NAAT and LPA, a second NAAT is performed at the Culture and Drug Susceptibility Test (C & DST) laboratory using the decontaminated deposit, and the microbiologist will provide the final decision. Additionally, direct LPA can be performed only on smear-positive specimens. In instances where the smear is negative, culture is set up, and if the culture is positive, an indirect LPA is performed on the isolate. 

Xpert Ultra is increasingly being used in the Private sector. Machine printouts of results from Xpert Ultra may be considered for guiding treatment; however, uncertainty in trace calls warrants clinical correlation / additional tests to be ordered by the treating physician.

 

2.4. Skin Test and Interferon-Gamma release assays (IGRAs) and Other Tests for TB

Tuberculin Skin Test (TST) is an intradermal injection of Purified Protein Derivative (PPD). (See Annexure 6 for TST technique) The current recommendation is to use 2TU PPD RT23 for all diagnostic purposes.  Mantoux’s test or PPD skin test is considered positive if the induration is 10 mm or more. In HIV co-infected cases, 5 mm may be taken as the cut-off. It is an immunological test that elicits delayed-type hypersensitivity. TST is used as an adjunct to other tests, and in the diagnosis of TB, it has the same connotation as a history of contact. A positive test indicates present or past infection with M.tb.  but cannot distinguish infection from disease. Currently, this test is not possible due to the non-availability of any form of the original lot of RT23 PPD. Most commercially available products are not from the standard mother lot and thus can cause problems in interpretation as the cut-offs derived for the standard product may not be applicable (Annexure 6). The utility of skin tests to assist the diagnosis of TB has decreased as the availability and sensitivity of microbiological tests has improved. 

Interferon-Gamma Release Assays (IGRAs), which provide an in vitro measure of M.tb hypersensitivity, offer the same information as TST and the tests currently approved by the WHO are TB Quantiferon Gold™ and TB Spot™ are available. Unlike Tuberculin Skin Test, they do not require a repeat visit for test reading and do not cross-react with BCG vaccination but need to be rapidly transferred to the lab and are expensive. A positive test marks the presence of TB infection in an individual but cannot confirm the presence of disease.

Diagnostics like serology (IgM, IgG, IgA antibodies against M. tb antigens), various in-house or non-validated commercial PCR tests and Bacillus Calmette–Guérin (BCG) tests are not recommended as TB diagnostic tools as they are often inaccurate. ESR is another test often used as a supportive investigation. Still, it is of no value in ruling in or ruling out the diagnosis of TB due to its non-specificity and the possibility to be affected by many variables other than the disease. A positive test result by either of the two available methods is not a reliable indicator that the person will progress to TB disease as the possibility of false-positive results cannot be ruled out. Conversely, a negative test result does not rule out TBI, given the case of a false-negative test result among at-risk groups, such as young children or among those recently infected.

C-Tb (Statens Serum Institut, Copenhagen, Denmark) is the next-generation skin test for the detection of TBI, which SSI Copenhagen has developed. It is an easy-to-use point-of-care test. It can deliver IGRA-like performance in a skin test format and uses a universal 5 mm cut off to differentiate the infected from the uninfected. The test is based on ESAT-6 and CFP-10 antigens (same as those used in IGRA) specific for M. tb and is unaffected by BCG vaccination. However, it is currently unavailable commercially and will be adopted in the programme once available and approved.