New advances in diagnosis in pulmonary tuberculosis

Diagnosis of Tuberculosis in children is challenging. Major challenge is because of difficulty in obtaining representative airway samples and paucibacillary nature of illness. Significant improvement occurred over past few years. However, the sensitivity and specificity of laboratory tests performed in children is less than desirable. The various methods for diagnosis of pulmonary TB in children includes screening by clinical features, demonstration/isolation of Mycobacterium tuberculosis or its components, demonstration of host's response to exposure to MTB including tuberculin skin test (TST), CTB, IGRA based tests and Biomarkers. Other important diagnostic support includes imaging findings in CXR, CT, and USG. Specimen for demonstration of MTB includes expectorated sputum, gastric aspirate/gastric lavage (GA/GL), Bronchoalveolar lavage (BAL), stool, urine, fine needle aspiration cytology (FNAC) from peripheral nodes, sampling from mediastinal nodes by EBUS/EUS. The methods used to demonstrate MTB include zeal neelson stain, Cartridge based nucleic acid amplification (CBNAAT), Xpert ultra, MGIT etc. Screening by Clinical features: As the clinical symptoms are nonspecific, therefore diagnosis based on clinical features alone has low specificity and is not recommended in clinical practice for diagnosis of pulmonary TB. Demonstration of mycobacterium tuberculosis (MTB) or its component MTB can be demonstrated in expectorated sputum in children above 8–10 years of age. In younger children GA/GL are commonly used method. Overnight hospitalization may be needed for GA/GL Studies have documented that ambulatory GA/GL have comparable results.1 Induced sputum is an emerging method of obtaining airway samples. Child is asked to come nil per orally for 6 h. Hypertonic saline 3/5% given by nebulizer after priming with salbutamol. Sputum/Nasopharyngeal secretion collected directly in sterile container and subjected to various investigations including smear, molecular tests and culture. It can be performed on an ambulatory basis in all age groups, including infants. Cartridge based nucleic amplification test on gastric aspirate showed variable positivity in children with pulmonary TB. A recently published study explored possible reasons for variable results. In a cohort study involving 394 children with presumptive pulmonary TB, Xpert-MTB/RIF on gastric aspirate improved diagnosis by twofold if the X-ray film showed lung parenchymal lesions as compared to those with hilar lymphadenopathy without active lung parenchymal lesions.2 A CBNAAT based test called Trunaat that battery operated and easy to perform in peripheral health care facilities was validated and compared with GeneXpert and found to have comparable results of both tests. Stool samples are easy to obtain and theoretically should give yield of microbiologically confirmed TB comparable to GA/GL. Earlier studies using stool samples for Xpert and MGIT culture reported low sensitivity. A recently published study testing stool by Xpert-ultra in 373 children with presumptive intrathoracic TB documented very good concordance between GA and stool positivity with a kappa of 0.83. Investigators also observe more Xpert Ultra positive results on stool (n = 27 (7.2%) than on sputum/GA (n = 23 (6.2%)).3 There is scope to improve it further GeneXpert Ultra in stool samples. Another easy to obtain sample in children is urine. Urine was tested for lipoarabinomannan antigen (LAM) of mycobacteria in 61 children with presumed TB. The urinary LAM level was higher in subjects with TB (1.80 + 1.02) mg/L compared to non-TB group (0.46 + 0.3) mg/L. The study calculated sensitivity and specificity of 83% and 85% respectively with a cut off value of 0.98 mg/L. Easy to perform and has good potential in pediatric TB. There is need to improve techniques to improve sensitivity and specificity.4 Small molecule bio-signature in childhood intra-thoracic tuberculosis. A recently published study compared serum metabolic profile using proton NMR spectroscopy of 23 children with culture-confirmed intra-thoracic TB and 13 non-TB controls. The study observed sensitivity, specificity and area under curve were 78.2%, 84.6%, and 0.86, respectively.5 If similar results are observed in future studies, this may hold promise in diagnosis of pediatric intrathoracic TB. CTB test: This is recently developed and validated test for identification of TB infection. The principle and method of performing the test is similar to that of Tuberculin skin test (TST). As the antigen included in CTB test are specific antigen for MTB (ESAT- 6 and CFP – 10 antigens), the specificity is better. There is potential of its use for identification of Tb infection for TPT. Line probe assay (LPA) is being performed for first line and second line drug sensitivity test (DST). As of now performed on smear positive specimen, report is available in 3–4 days. It can be performed on MTB obtained on MGIT culture for DST. Bronchoscopy and broncho alveolar lavage (BAL): Obtaining image guided bronchoalveolar lavage and testing by GeneXpert had shown to increase yield of microbiologically confirmed TB in children who had pulmonary TB but GeneXpert was negative on GA/GL. IGRA based test (QFT/Ellispot) These tests are being used as an adjunct for diagnosis of TB and for identification of children for treatment for prevention of TB (TPT). In countries with high prevalence of TB, it is considered to be comparable to TST. Now CTB may be considered as an alternative to IGRA based test. CXR in TB CXR findings that suggest TB disease include mediastinal lymphadenopathy: hilar, paratracheal, subcarinal, miliary shadows and cavities. These are suggestive but not confirmatory findings and require investigations to confirm diagnosis of TB disease. For nonspecific infiltrates it is suggested to give a course of antibiotics and repeat CXR after 2–4 weeks. If CXR abnormalities persist, a repeat GA/GL may be done before starting treatment for TB. CT chest in Intrathoracic TB Contrast enhanced thoracic CT scan in children with mediastinal nodes may show conglomerate necrotic lymph nodes with rim enhancement and obscuration of perinodal fat. Thoracic CT scan may help in targeted sampling by identification of nodes and affected segment of lungs. S K Kabra: Conceptualization; Methodology; Writing—original draft; Writing—review and editing. The author declares no conflicts of interest. This is a review article. The article does not include data on biomedical research.