3.1 Active tuberculosis

Select language:
On this page

    3.1.1 Introduction

    Active tuberculosis (TB) is bacteriologically confirmed by the detection of M. tuberculosis complex through different bacteriological tests. These tests detect either the organism (smear microscopy and culture), or some of its genetic material (genotypic tests, including rapid molecular tests and genome sequencing).

     

    Specimens used for bacteriological testing include respiratory specimens (sputum, nasopharyngeal aspirate and, in children, gastric aspirate) and extrapulmonary specimens (Table 3.6).

     

    Drug susceptibility testing (DST) is indicated for all patients with confirmed TB. It can be performed using genotypic or phenotypic tests:

    • Genotypic DST (gDST) can detect resistance to TB drugs by identifying specific gene mutations.
    • Phenotypic DST (pDST) can detect resistance to TB drugs by measuring the growth of M. tuberculosis in the presence of the drug.

     

    To diagnose TB and determine the appropriate regimen at baseline:

    • All patients should be tested with a rapid molecular test (RMT) to detect M. tuberculosis and rifampicin resistance a Citation a. When microscopy is the only diagnostic test available, specimens should be sent to a facility with capacity to perform RMTs. .
    • Whatever the result of the rifampicin susceptibility test (resistance detected or not), all patients, if possible, should be tested with an RMT for isoniazid resistance and at least those with high risk of isoniazid resistance (for definition of high risk of resistance, see below).
    • All patients with rifampicin resistance should be tested for resistance to fluoroquinolones and other TB drugs.
    • All patients with isoniazid resistance and rifampicin susceptibility should be tested for resistance to fluoroquinolones.
    • Culture, pDST and genome sequencing may be required.

     

    In limited-resource settings, resistance to TB drugs should be investigated in priority in patients with:

    • High risk of mortality: e.g. HIV-infected patients or patients with extensive disease.
    • High risk of resistance: patients with previous TB treatment, or in contact with a TB case resistant to TB drug(s), or coming from an area of high prevalence of resistance to TB drug(s).

     

    Notes:

    • Negative bacteriological tests for M. tuberculosis does not rule out TB.
    • A negative DST does not necessarily rule out drug resistance.

     

    Other investigations can assist TB diagnosis. These investigations include: lateral flow urine lipoarabinomannan assay (LF-LAM) which detects an antigen of M. tuberculosis cell wall excreted in urine, medical imaging, and some biological tests.

    3.1.2 Rapid molecular tests

    RMTs are nucleic acid amplification tests (NAATs). They can detect M. tuberculosis and drug resistance by identifying resistance-conferring mutations in certain genes (Table 3.1). Other drug resistance- conferring mutations may be present, but not detected by RMTs. In areas where prevalence of these mutations is high, RMT sensitivity may be decreased [1] Citation 1. Sanchez-Padilla E, Merker M, Beckert P, Jochims F, Dlamini T, Kahn P, Bonnet M, Niemann S. Detection of drug-resistant tuberculosis by Xpert MTB/RIF in Swaziland. N Engl J Med. 2015 Mar 19;372(12):1181-2. https://www.nejm.org/doi/10.1056/NEJMc1413930?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%20%200www.ncbi.nlm.nih.gov .

     

    Table 3.1  Rapid molecular tests and detection of drug resistance

     

    Tests

    TB drug resistance (targeted genes)

    Low complexity NAATs

    • Xpert MTB/RIF
    • Xpert MTB/RIF Ultra
    • Truenat MTB-RIF Dx

    Moderate complexity NAATs

    High complexity NAATs

    • GenoType MTBDRplus (V2.0)
    • Genoscholar NTM+MDRTB II

    Rifampicin (rpoB)

    Low complexity NAATs

    • Xpert MTB/XDR

    Moderate complexity NAATs

    High complexity NAATs

    • GenoType MTBDRplus (V2.0)
    • Genoscholar NTM+MDRTB II

    Isoniazid high-level resistance (katG)

    Isoniazid low-level resistance, thionamides (a) Citation a. Mutations in other genes can result in resistance to thionamides. Consequently, absence of inhA mutation does not rule out resistance.  (inhA promoter)

    Low complexity NAATs

    • Xpert MTB/XDR

    High complexity NAATs

    • GenoType MTBDRsl (V2.0)

    Fluoroquinolones (gyrA, gyrB) (b) Citation b. Specific mutations in gyrA (e.g. mutations recognized by the probes MUT3B, 3C, 3D) are associated with high-level fluoroquinolones resistance.
    RMTs have a good specificity, but are less sensitive than culture. Their various levels of complexity determine their use at different levels of health facilities. Low complexity RMTs are preferred in routine practice.

    Aminoglycosides (rrs, eis)

    High complexity NAATs

    • Genoscholar PZA-TB II

    Pyrazinamide (pncA)

    RMTs have a good specificity, but are less sensitive than culture. Their various levels of complexity determine their use at different levels of health facilities. Low complexity RMTs are preferred in routine practice.

    Low complexity nucleic acid amplification tests

    1) Xpert assays  

    Xpert assays are almost fully automated. An uninterrupted power supply and a computer are required to perform and read assays.

     

    Xpert assays can be performed on:

    • Respiratory specimens
    • Extrapulmonary (EP) specimens:
      • Lymph node biopsy or aspirate: suspicion of lymph node TB or detection of rifampicin resistance in clinically diagnosed lymph node TB;
      • Cerebrospinal fluid (CSF): suspicion of TB meningitis;
      • Pleural fluid: suspicion of TB with pleural effusion;
      • Stool: suspicion of PTB in children;
      • Pericardial fluid: suspicion of TB with pericardial effusion (sampling to be performed only by experienced clinicians);
      • Urine: suspicion of genitourinary TB; suspicion of disseminated TB in HIV-infected patients;
      • Synovial fluid: suspicion of TB arthritis;
      • Peritoneal fluid: suspicion of abdominal TB;
      • Blood:  suspicion of disseminated TB in HIV-infected patients.

     

    Xpert MTB/RIF and Xpert MTB/RIF Ultra assays provide simultaneously results for M. tuberculosis detection and rifampicin resistance.

     

    Sensitivity of Xpert MTB/RIF Ultra assay is higher than that of Xpert MTB/RIF assay. It provides a result “trace” corresponding to the lowest bacillary load for M. tuberculosis detection. It is preferred for HIV-infected patients, children, EP specimens, and sputum smear-negative specimens. Its specificity is lower in patients with a history of TB, as a “trace” result may indicate that the specimen contains fragments of dead bacilli.  

     

    WHO has validated their use on lymph node biopsy or aspirate, CSF, and pleural fluid, synovial fluid [2] Citation 2. World Health Organization. WHO operational handbook on tuberculosis. Module 3: Diagnosis - rapid diagnostics for tuberculosis detection. Geneva: World Health Organization; 2020.
    https://apps.who.int/iris/rest/bitstreams/1284635/retrieve
    . Xpert assays can be performed on any biopsy specimens (lymph node, bone, skin, resection material, etc.) with good performance. Xpert assays have shown acceptable performances in various studies on other specimens (peritoneal [3] Citation 3. Rufai SB, Singh S, Singh A, Kumar P, Singh J, Vishal A. Performance of Xpert MTB/RIF on Ascitic Fluid Samples for Detection of Abdominal Tuberculosis. J Lab Physicians. 2017;9(1):47-52.
    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5015498/
     and pericardial [4] Citation 4. Andrianto A, Mertaniasih NM, Gandi P, et al. Diagnostic test accuracy of Xpert MTB/RIF for tuberculous pericarditis: a systematic review and meta-analysis. F1000Res. 2020;9:761. Published 2020 Jul 22.
    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7417956/
    fluids, stools [5] Citation 5. Walters E, van der Zalm MM, Palmer M, et al. Xpert MTB/RIF on Stool Is Useful for the Rapid Diagnosis of Tuberculosis in Young Children With Severe Pulmonary Disease. Pediatr Infect Dis J. 2017;36(9):837-843. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5558052/ [6] Citation 6. MacLean E, Sulis G, Denkinger CM, Johnston JC, Pai M, Ahmad Khan F. 2019. Diagnostic accuracy of stool Xpert MTB/RIF for detection of pulmonary tuberculosis in children: a systematic review and meta-analysis. J Clin Microbiol 57:e02057-18.
    https://doi.org/10.1128/JCM.02057-18
    [7] Citation 7. World Health Organization. WHO consolidated guidelines on tuberculosis. Module 3: Diagnosis - rapid diagnostics for tuberculosis detection, 2021 update. Geneva 2021.
    https://www.who.int/publications/i/item/9789240029415
    , and urine [8] Citation 8. Chen, Y., Wu, P., Fu, L. et al. Multicentre evaluation of Xpert MTB/RIF assay in detecting urinary tract tuberculosis with urine samplesSci Rep 9, 11053 (2019).
    https://doi.org/10.1038/s41598-019-47358-3
    [9] Citation 9. Peter JG, Theron G, Muchinga TE, Govender U, Dheda K. The diagnostic accuracy of urine-based Xpert MTB/RIF in HIV-infected hospitalized patients who are smear-negative or sputum scarce. PLoS One. 2012;7(7):e39966.
    https://pubmed.ncbi.nlm.nih.gov/22815718/
    ). Xpert assays on blood have a low sensitivity compared to culture and are not routinely recommended [10] Citation 10. Pohl C, Rutaihwa LK, Haraka F, Nsubuga M, Aloi F, Ntinginya NE, Mapamba D, Heinrich N, Hoelscher M, Marais BJ, Jugheli L, Reither K. Limited value of whole blood Xpert(®) MTB/RIF for diagnosing tuberculosis in children. J Infect. 2016 Oct;73(4):326-35. doi: 10.1016/j.jinf.2016.04.041. Epub 2016 Jul 7. PMID: 27394403.
    https://www.journalofinfection.com/article/S0163-4453(16)30161-X/fulltext
    .

     

    Xpert MTB/XDR assay detects resistance to isoniazid (low- and high-level), fluoroquinolones (low- and high-level), aminoglycosides, and thionamides. It does not detect resistance to rifampicin.

     

    Xpert MTB/XDR assay employ the same platform as other Xpert assays, but require a 10-colour module instead of the 6-colour module used for Xpert MTB/RIF and Xpert MTB/RIF Ultra assays. The 10-colour module can also read Xpert MTB/RIF and Xpert MTB/RIF Ultra assays.

     

    Xpert MTB/XDR assay should be used:

    • When resistance to rifampicin has been detected by Xpert MTB/RIF or Xpert MTB/RIF Ultra, to detect resistance to other drugs.
    • When M. tuberculosis has been detected by Xpert MTB/RIF or Xpert MTB/RIF Ultra or culture, to detect resistance to isoniazid in all patients, if possible, and at least those with high risk of isoniazid resistance (Section 3.1.1).
    • Before using a fluoroquinolone containing regimen in isoniazid-resistant TB (Hr-TB), multidrug-resistant (MDR-TB), rifampicin-resistant TB (RR-TB) or drug-susceptible TB treated with the regimen 2HPZ-Mfx/2HP-Mfx.
    • Before treating drug-susceptible TB meningitis with the regimen 6HRZ-Eto.
    • In patients with fluoroquinolone-susceptible TB, initially treated with a fluoroquinolone-containing regimen, and presenting a smear-positive microscopy at Month 2 or later.

     

    Table 3.2 – Main performances of Xpert assays

     

    Xpert

    assays

    Performances

     MTB/RIF

    Detection of M. tuberculosis (MTB) compared to culture:

    • Sensitivity in respiratory specimens [11] Citation 11. World Health Organization. Automated real-time nucleic acid amplification technology for rapid and simultaneous detection of tuberculosis and rifampicin resistance: Xpert MTB/RIF assay for the diagnosis of pulmonary and extrapulmonary TB in adults and children. Policy update. Geneva 2013.
      https://apps.who.int/iris/bitstream/handle/10665/112472/9789241506335_eng.pdf?sequence=1
      :
      • sputum-smear positive: 99%
      • sputum smear-negative: 68%
      • HIV-infected patients: 79%
      • children: see Appendix 1.
    • Sensitivity in EP specimen: see Appendix 1.
    • Specificity: very high in all specimens (99%), i.e. a positive result is unlikely to be a false positive.

    Detection of rifampicin resistance compared to pDST [11] Citation 11. World Health Organization. Automated real-time nucleic acid amplification technology for rapid and simultaneous detection of tuberculosis and rifampicin resistance: Xpert MTB/RIF assay for the diagnosis of pulmonary and extrapulmonary TB in adults and children. Policy update. Geneva 2013.
    https://apps.who.int/iris/bitstream/handle/10665/112472/9789241506335_eng.pdf?sequence=1
    :

    Sensitivity: 95%; specificity: 98%

     MTB/RIF Ultra

    Detection of MTB in respiratory and EP specimens [12] Citation 12. World Health Organization. WHO meeting report of a technical expert consultation: non-inferiority analysis of Xpert MTF/RIF Ultra compared to Xpert MTB/RIF. Geneva 2017.
    https://apps.who.int/iris/bitstream/handle/10665/254792/WHO-HTM-TB-2017.04-eng.pdf?sequence=1
    :

    • Sensitivity: + 5% compared to Xpert MTB/RIF
    • Specificity: - 3.2% compared to Xpert MTB/RIF; - 5.4% in patients with a history of TB

    No result for rifampicin resistance if “trace” result.

    MTB/XDR

    Detection of MTB in respiratory and EP specimens (children and adults):

    As Xpert MTB/RIF.

    Detection of resistances compared to pDST [13] Citation 13. World Health Organization. Update on the use of nucleic acid amplification tests to detect TB and drug-resistant TB: rapid communication. Geneva: World Health Organization; 2021.
    https://apps.who.int/iris/rest/bitstreams/1332438/retrieve
    :

    • To isoniazid (low- and high-level): sensitivity: 94.2%; specificity: 98%
    • To fluoroquinolones (low- and high-level): sensitivity: 93.1%; specificity: 98.3%
    • To aminoglycosides: sensitivity: 86.1%; specificity: 98.9%
    • To thionamides: sensitivity: 51.7%; specificity: 98.3%

     

    For more information on specimen processing and Xpert instruments see Appendix 1.

    For interpretation of Xpert assay results see Appendix 2.

    2) Truenat assays

    Truenat assays require:

    • Several manual steps (pipetting).
    • Sequential testing for M. tuberculosis detection (Truenat MTB Plus), then for rifampicin resistance detection (Truenat MTB-RIF Dx).
    • Separate kits for specimen preparation, DNA extraction, DNA amplification, and detection of M. tuberculosis and rifampicin resistance.

     

    Truenat MTB Plus can only be performed on sputum specimens (positive or negative smear microscopy). It is not recommended for other respiratory specimens or EP specimens [2] Citation 2. World Health Organization. WHO operational handbook on tuberculosis. Module 3: Diagnosis - rapid diagnostics for tuberculosis detection. Geneva: World Health Organization; 2020.
    https://apps.who.int/iris/rest/bitstreams/1284635/retrieve
    [7] Citation 7. World Health Organization. WHO consolidated guidelines on tuberculosis. Module 3: Diagnosis - rapid diagnostics for tuberculosis detection, 2021 update. Geneva 2021.
    https://www.who.int/publications/i/item/9789240029415
    .

    Specificity is high, i.e. a positive result is unlikely to be a false positive [14] Citation 14. StopTB Partnership. Practical Guide to Implementation of Truenat Tests for the Detection of TB and Rifampicin Resistance. Geneva 2021.
    http://stoptb.org/assets/documents/resources/publications/sd/Truenat_Implementation_Guide.pdf
    .

    Tests can be run at room temperatures of up to 40 °C and humidity of up to 80%. Truenat instruments are battery-operated and can be used in peripheral or mobile health facilities.

    Interpretation of results is the same as for Xpert (Appendix 2).

     

    Table 3.3 – Main performances of Truenat assays

     

    Truenat

    assays

    Performances

    MTB Plus

    Detection of MTB in sputum specimens (children and adults) compared to culture:  

    MTB-RIF Dx

    Detection of rifampicin resistance compared to pDST:

    Performed on the DNA isolated from sputum specimens with Truenat MTB Plus positive result. 

    • Sensitivity: 84%
    • Specificity: 97%

     

    3) TB-LAMP

    Although validated by WHO, this test has major limitations:

    • It does not detect rifampicin resistance.
    • Its sensitivity is lower than that of other low complexity NAATs in HIV-infected or smear-negative patients.
    • It cannot be used for the diagnosis of extrapulmonary TB (EPTB) [2] Citation 2. World Health Organization. WHO operational handbook on tuberculosis. Module 3: Diagnosis - rapid diagnostics for tuberculosis detection. Geneva: World Health Organization; 2020.
      https://apps.who.int/iris/rest/bitstreams/1284635/retrieve
      .

     

    Box 3.1 – Choice of low complexity NAATs

     

    Xpert: first line tests for the diagnosis of TB in children and adults.

    Truenat: if no power supply or operating temperature between 31 and 40 °C.

    TB-LAMP: not recommended.

     

    Moderate complexity nucleic acid amplification tests

    WHO recommends these tests for the simultaneous detection of M. tuberculosis and resistance to rifampicin and isoniazid, from smear-positive and negative respiratory specimens, in children and adults, including HIV-infected patients.

     

    Table 3.4 – Performances of moderate complexity NAATs

     

    Tests

    Performances

    • Abbott Real Time MTB and MTB RIF/INH BD MAX MDR-TB
    • Hain FluoroType MTB and MTBDR
    • Roche cobas MTB and MTB-INH/RIF

    Detection of MTB compared to culture:

    • Sensitivity 93%
    • Specificity 97.7%

    Detection of rifampicin resistance compared to pDST:

    • Sensitivity 96.7%
    • Specificity 98.9%

    Detection of isoniazid resistance compared to pDST:

    • Sensitivity 86.4%
    • Specificity 99.8%

     

    NAATs of moderate complexity have several limitations:

    • Need for space, equipment, qualified staff; only feasible in regional laboratories.
    • Their use does not eliminate the need for pDST, high complexity NAATs, or genome sequencing to:
      • test susceptibility to other TB drugs;
      • confirm a negative result in patients at high risk of drug resistance.
    • Their use on EP specimens is not validated.

    High complexity nucleic acid amplification tests

    Line probe assays (LPA) can detect specific rifampicin, isoniazid, fluoroquinolones, aminoglycosides, and pyrazinamide resistance encoding mutations in M. tuberculosis.

     

    These tests can be performed on isolates of M. tuberculosis (indirect testing). Some can be performed on sputum specimens (direct testing).

     

    NAATs of high complexity have several limitations:

    • Need for space, equipment, highly qualified staff; only feasible in reference and national laboratories.
    • Risk of cross-contamination (tests are performed in an open system that can lead to the detection of DNA from sources other than the specimen).
    • To benefit from the short turnaround time of these tests, efficient logistical support is required to ensure specimens are transported to the laboratory and the results are delivered in a timely manner.
    • Their use does not eliminate the need for pDST or genome sequencing to:
      • test sensitivity to other TB drugs;
      • confirm a negative result in patients at high risk of drug resistance.
    • Although direct test results can be obtained in 1 to 2 days, for indirect tests, it is necessary to wait the time required for bacterial growth (Appendix 5).
    • Their use on respiratory (non-sputum) or EP specimens is not validated.

     

    Box 3.2 – WHO validated LPAs b Citation b. For more information, see: Global Laboratory Initiative. Line probe assays for drug resistant tuberculosis detection Interpretation and reporting guide for laboratory staff and clinicians. http://stoptb.org/wg/gli/assets/documents/LPA_test_web_ready.pdf

     

    First-line LPAs

    • GenoType MTBDRplus version 2 (“Hain first line test”): initial test to detect resistance to rifampicin and isoniazid on smear-positive sputum specimens and M. tuberculosis isolates. Compared to pDST, sensitivity is 98.2% for rifampicin, and 97.8% for isoniazid; specificity is 95.4% for rifampicin, and 98.8% for isoniazid [15] Citation 15. World Health Organization. WHO Guideline: The use of molecular line probe assays for the detection of resistance to isoniazid and rifampicin. 21–23 (2016).
      https://apps.who.int/iris/bitstream/handle/10665/246131/9789241510561-eng.pdf?sequence=1&isAllowed=y
      . On smear-negative sputum specimens, sensitivity is low (44.4%), and its use is not recommended [13] Citation 13. World Health Organization. Update on the use of nucleic acid amplification tests to detect TB and drug-resistant TB: rapid communication. Geneva: World Health Organization; 2021.
      https://apps.who.int/iris/rest/bitstreams/1332438/retrieve
      .
    • Genoscholar NTM+MDRTB II (“Nipro test”): performances comparable to GenoType MTBDRplus to detect resistance to rifampicin and isoniazid on smear-positive sputum specimens and M. tuberculosis isolates. Not recommended on smear-negative sputum specimens. Can differentiate M. avium, M. intracellulare and M. kansasii from other non-tuberculous mycobacteria.
    • Genoscholar PZA-TB ll: to detect resistance to pyrazinamide on M. tuberculosis isolates. Compared to pDST, sensitivity is 81%, and specificity is 97% [13] Citation 13. World Health Organization. Update on the use of nucleic acid amplification tests to detect TB and drug-resistant TB: rapid communication. Geneva: World Health Organization; 2021.
      https://apps.who.int/iris/rest/bitstreams/1332438/retrieve
      .

     

    Second-line LPA

    GenoType MTBDRsl version 2 (“Hain second line test”): in patients with confirmed MDR/RR-TB, to detect resistance to fluoroquinolones (high- and low-level) and aminoglycosides on smear-positive or smear-negative sputum specimens and M. tuberculosis isolates. The number of “indeterminate” results is higher for smear-negative than for smear-positive sputum specimens. For smear-positive sputum specimens, sensitivity is 93% for fluoroquinolones, and 88.9% for aminoglycosides; specificity is 98.3% for fluoroquinolones, and 91.7% for aminoglycosides [16] Citation 16. Tagliani, E. et al. Diagnostic performance of the new version (v2.0) of GenoType MTBDRsl assay for detection of resistance to fluoroquinolones and second-line injectable drugs: A multicenter study. J. Clin. Microbiol. 53, 2961–2969 (2015).
    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4540937/pdf/zjm2961.pdf
    .

     

    3.1.3 Genome sequencing

    Genome sequencing can only be performed in highly specialized reference laboratories.

    It can rapidly:

    • Detect mutations associated with TB drug resistance. When available, it is particularly useful to identify:  
      • resistance to TB drugs for which pDST is unreliable, or no RMTs are available;
      • mutations missed by RMTs (+ 20% of drug resistance detection compared to RMTs has been described [17] Citation 17. Connie Lam, Elena Martinez, Taryn Crighton, Catriona Furlong, Ellen Donnan, Ben J. Marais, Vitali Sintchenko, Value of routine whole genome sequencing for Mycobacterium tuberculosis drug resistance detection. International Journal of Infectious Diseases, 2021.
        https://www.ijidonline.com/article/S1201-9712(21)00251-4/fulltext
        ).
    • Detect mixed infection (infection with distinct M. tuberculosis strains).
    • Identify heteroresistance (same strain, with different resistance profiles).
    • Differentiate treatment relapse and reinfection with a different strain.

     

    Genome sequencing methods include Sanger sequencing (reference method) and next generation sequencing (NGS). The advantage of NGS is that, unlike Sanger sequencing, it provides results for a large number of genes in a single reaction.

    NGS results are interpreted by reference laboratories using specific software and mutation databases c Citation c. For more information :
    • WHO catalogue of mutations in M. tuberculosis complex and their association with drug resistance: https://www.who.int/publications/i/item/9789240028173
    • Relational Sequencing TB (ReSeqTB) Data Platform: https://c-path.org/programs/cptr/cptr-tools/databases/relational-sequencing-tb-data-platform-reseqtb/
    .

    Some mutations associated with resistance to recently introduced drugs (e.g. bedaquiline and delamanid) and their therapeutic implications are still not well-known.

     

    The two main NGS techniques are targeted NGS (tNGS) and whole genome sequencing (WGS):

    • tNGS (on smear-positive sputum specimens or culture isolates): detection of resistance conferring mutations on 18 selected genes: first-line TB drugs, fluoroquinolones, aminoglycosides, linezolid, bedaquiline, clofazimine, ethionamide (Deeplex®Myc-TB). Used in routine.
    • WGS (on culture isolates): detection of resistance conferring mutations on whole genome (i.e. potentially all TB drugs). Used for research.

    3.1.4 Smear microscopy

    The purpose of smear microscopy is to detect acid-fast bacilli (AFB) in stained specimens.

     

    Smear microscopy has several limitations:  

    • It has a sensitivity lower than RMTs and culture in respiratory specimens (65% compared to culture [11] Citation 11. World Health Organization. Automated real-time nucleic acid amplification technology for rapid and simultaneous detection of tuberculosis and rifampicin resistance: Xpert MTB/RIF assay for the diagnosis of pulmonary and extrapulmonary TB in adults and children. Policy update. Geneva 2013.
      https://apps.who.int/iris/bitstream/handle/10665/112472/9789241506335_eng.pdf?sequence=1
      ) and EP specimens (48% compared to culture [18] Citation 18. Enrico Tortoli, Cristina Russo, Claudio Piersimoni, Ester Mazzola, Paola Dal monte, Michela Pascarella, Emanuele Borroni, Alessandra Mondo, Federica Piana, Claudio Scarparo, Luana Coltella, Giulia Lombardi, Daniela M. Cirillo. Clinical validation of Xpert MTB/RIF for the diagnosis of extrapulmonary tuberculosis. European Respiratory Journal 2012 40: 442-447.
      https://erj.ersjournals.com/content/40/2/442
      ).
    • It has a low sensitivity in patients with low bacillary load in sputum (paucibacillary TB), e.g. children and HIV-infected patients. 
    • It cannot differentiate between M. tuberculosis and non-tuberculous mycobacteria. However, in areas with high TB prevalence, AFB detected on smear microscopy are most likely M. tuberculosis.
    • It does not determine if bacilli are viable (alive) or non-viable (dead).
    • It does not determine susceptibility of the bacilli to TB drugs.

     

    Sputum smear microscopy is no longer the recommended initial diagnostic test for PTB. However, it still plays a role:

    • When RMTs are not available.
    • For assessing the infectiousness of PTB patients.
    • For monitoring the response to TB treatment in patients with:
      • drug-susceptible PTB (Chapter 9)
      • drug-resistant PTB. However, culture is also required for monitoring treatment response in these patients (Chapter 10 and Chapter 11).

     

    For improving the sensitivity of smear microscopy:

    1) Two sputum specimens should be examined. Approximately 86% of sputum smear-positive patients are identified during the first examination, and an additional 12% during the second. It is not necessary to carry out more than 2 examinations [19] Citation 19. Mase, S.R., et al. Yield of serial sputum specimen examinations in the diagnosis of pulmonary tuberculosis: a systematic review. Int J Tuberc Lung Dis, 2007. 11(5): p. 485-95.
    http://docserver.ingentaconnect.com/deliver/connect/iuatld/10273719/v11n5/s3.pdf?expires=1611823630&id=0000&titleid=3764&checksum=39399CE1057042BD71BAC51B1470C1F8
    .

    2) Light-emitting diode (LED) fluorescent microscopy to examine auramine-stained smears is preferred to Ziehl-Neelsen microscopy, as it is more sensitive, and reading is more rapid.

     

    Concentration techniques can also increase the sensitivity of smear microscopy [20] Citation 20. Bonnet M, Ramsay A, Githui W, Gagnidze L, Varaine F, Guerin PJ. Bleach Sedimentation: An Opportunity to Optimize Smear Microscopy for Tuberculosis Diagnosis in Settings of High Prevalence of HIV. Clin Infect Dis. 2008 Jun.;46(11):1710–6. .

     

    For sputum specimen collection, storage and shipment see Appendix 3.

    For sputum smear preparation and staining techniques see Appendix 4.

    3.1.5 Culture

    Culture consists of growing M. tuberculosis in specific liquid or solid media.

    Culture on liquid medium (automated or manual mycobacterial growth indicator tube, MGIT) is the reference method for the diagnosis of PTB and EPTB. Given the long turnaround time and equipment required, it is not used as initial diagnostic test.

    Culture on solid medium (Lowenstein-Jensen) is cheaper, less prone to contaminations than cultures on liquid media, but its turnaround time is longer.  

    Other culture techniques are less commonly used d Citation d. Microscopic observation of drug susceptibility (MODS), nitrate reductase assay (NRA), thin layer agar and colorimetric redox indicator (CRI). .

     

    Culture is necessary to:

    • Confirm treatment failure.
    • Assess treatment response in patients with drug-resistant PTB (Chapter 10 and Chapter 11).
    • Evaluate treatment outcome in patients with drug-resistant PTB (Chapter 17).
    • Provide isolates for the following tests:
      • First-line LPAs on sputum smear-negative and EP specimens
      • Genoscholar PZA-TB ll, regardless of sputum smear positivity
      • First- or second-line LPA when an initial direct LPA gives an invalid result
      • WGS
      • tNGS on smear-negative sputum specimens
    • Differentiate between M. tuberculosis and non-tuberculosis mycobacteria. Differentiation between species within the M. tuberculosis complex is not routinely performed.

     

    Culture may help to diagnose TB when other bacteriological tests are negative or inconclusive:

    • In patients with signs and symptoms of TB and a negative RMT, particularly when resistance is suspected.
    • In adults with history of TB in the previous 5 years and showing a “trace” result by Xpert MTB/RIF Ultra.

     

    Culture has several limitations:

    • Only specialized laboratories implementing systematic quality assurance procedures can be relied upon for culture (often national reference laboratories or supranational).
    • M. tuberculosis is a slow-growing bacillus. Positive culture results are obtained after 2 to 4 weeks.

     

    For sputum specimen collection, storage and shipment see Appendix 3.

    For the time required to obtain the results see Appendix 5.

    3.1.6 Phenotypic drug susceptibility testing

    Phenotypic DST (pDST) determines if a strain is resistant to a TB drug by evaluating the growth in the presence of the drug.
    It can determine two levels of resistance (low and high) for isoniazid and fluoroquinolones.

     

    The pDST is essential to detect resistance to drugs for which there are no reliable RMTs, and when genome sequencing is not available.

    In addition, pDST may be necessary:

    • If an RMT indicates M. tuberculosis “detected” and drug resistance “indeterminate”.
    • If an RMT indicates drug susceptibility in a patient at high risk of resistance.
    • In areas with a high prevalence of mutations not detected by RMTs.

     

    Phenotypic DST is performed on culture isolates by specialized laboratories (often national reference laboratories or supranational).

     

    The pDST is not reliable for all drugs, even when performed by a highly qualified laboratory [21] Citation 21. Technical manual for drug susceptibility testing of medicines used in the treatment of tuberculosis.
    https://apps.who.int/iris/bitstream/handle/10665/275469/9789241514842-eng.pdf?ua=1
    .

     

    Table 3.5 – Reliability of pDST for first- and second-line TB drugs

     

    Reliability of pDST

    TB drugs

    Highly reliable

    Isoniazid

    Rifampicin

    Fluoroquinolones

    Aminoglycosides

    Unreliable (should not be performed)

    Ethambutol

    Ethionamide

    Cycloserine or terizidone

    Para-aminosalicylic acid (or sodium)

    Delamanid

    Reliable, but limited access outside of supranational laboratories

    Bedaquiline

    Linezolid

    Clofazimine

    Reliable when performed in a high-quality laboratory (difficult to perform)

    Pyrazinamide

     

    3.1.7 Summary of bacteriological tests

    The tables below provide an overview of the specimens that can be used for each test, and of the tests that can detect resistance to each TB drug.

     

    Table 3.6 – Specimens for bacteriological tests

     

    Tests

    Specimens

    Xpert, microscopy, culture

    Respiratory or EP specimens

    Truenat

    Sputum (smear-positive or negative)

    Moderate complexity NAATs

    Respiratory specimens

    GenoType MTBDRplus version 2

    Genoscholar NTM+MDRTB II

    Sputum (smear-positive only)

    M. tuberculosis isolate

    Genoscholar PZA-TB ll

    M. tuberculosis isolate

    GenoType MTBDRsl version 2

    Sputum (smear-positive or negative)

    M. tuberculosis isolate

    tNGS

    Sputum (smear-positive only)

    M. tuberculosis isolate

    WGS

    M. tuberculosis isolate

     

    Table 3.7 - Tests to detect specific drug resistance

     

    TB drugs

    gDST

    pDST

    Rifampicin

    Xpert MTB/RIF and Ultra, Truenat MTB-RIF Dx

    Moderate complexity NAATs

    GenoType MTBDRplus, Genoscholar NTM+MDRTB II

    Genome sequencing

    Yes

    Isoniazid (c) Citation c. High- and low-level resistance detected by gDST and pDST.

    Xpert MTB/XDR

    Moderate complexity NAATs

    GenoType MTBDRplus, Genoscholar NTM+MDRTB II

    Genome sequencing

    Yes

    Pyrazinamide

    Genoscholar PZA-TB ll

    Genome sequencing

    Yes

    Ethambutol

    Genome sequencing

    Unreliable

    Fluoroquinolones (c) Citation c. High- and low-level resistance detected by gDST and pDST.

    Xpert MTB/XDR

    GenoType MTBDRsl

    Genome sequencing

    Yes

    Amikacin

    Xpert MTB/XDR

    GenoType MTBDRsl

    Genome sequencing

    Yes
     

    Streptomycin

    Genome sequencing

    Yes (d) Citation d. Rarely available in resource-limited settings.

    Thionamides (e) Citation e. Most mutations conferring resistance to thionamides are not detected by RMTs.

    Xpert MTB/XDR

    GenoType MTBDRplus, Genoscholar NTM+MDRTB II

    Genome sequencing

    Unreliable

    Cycloserine or terizidone

    Para-aminosalicylic acid (or sodium)

    Whole genome sequencing

    Unreliable

    Bedaquiline
    Linezolid

    Clofazimine

    Delamanid

    Genome sequencing

    Yes (d) Citation d. Rarely available in resource-limited settings.

    3.1.8 Lateral flow urine lipoarabinomannan assay

    TB lipoarabinomannan (LF-LAM) is a urine-based point-of-care test that detects lipoarabinomannan (LAM) antigen e Citation e. LAM antigen is a component of the mycobacterial cell walls released by M. tuberculosis then excreted by the kidneys. , which is a marker of active TB.

     

    This test is easy to perform by trained staff, including in peripheral heath facilities.

     

    Advantages of LF-LAM over sputum-based tests include:

    • Urine specimens easier to collect.
    • No risk of staff contamination during specimen collection or processing.
    • No specific storage requirements for the urine prior to testing.

     

    The urine is applied to the test strip, left at room temperature for 25 minutes, then read by the naked eye by comparing the band for positivity to a grading scale provided by the manufacturer f Citation f. Alere Determine® TB LAM Ag (Alere Inc, Waltham, MA, USA). .

     

    This rapid test should be used in the diagnosis of PTB and EPTB in HIV-infected children and adults. Its rapidly obtained result can contribute to reducing TB mortality among these patients [7] Citation 7. World Health Organization. WHO consolidated guidelines on tuberculosis. Module 3: Diagnosis - rapid diagnostics for tuberculosis detection, 2021 update. Geneva 2021.
    https://www.who.int/publications/i/item/9789240029415
    .

     

    Its performances depend on the individual level of immunodeficiency at the time of testing. Its sensitivity is low, but it has an acceptable specificity (see below).

     

    The LF-LAM test is recommended for the following patient groups [22] Citation 22. World Health Organization. Lateral flow urine lipoarabinomannan assay (LF-LAM) for the diagnosis of active tuberculosis in people living with HIV. Policy update 2019. Geneva: World Health Organization; 2019.
    https://apps.who.int/iris/bitstream/handle/10665/329479/9789241550604-eng.pdf?sequence=1&isAllowed=y&ua=1
    :

    • HIV-infected patients with signs and symptoms of TB or seriously ill g Citation g. Seriously ill: respiratory rate > 30/minute, temperature > 39 °C, heart rate > 120/minute and unable to walk unaided. , irrespective of CD4 count (sensitivity: 35%; specificity: 95%).
    • Hospitalised patients with advanced HIV disease h Citation h. For children > 5 years and adults: CD4 count < 200 cells/mm3 or a WHO clinical stage 3 or 4. All children < 5 years are considered as having advanced HIV disease. (sensitivity: 64%; specificity: 82%).
    • HIV-infected outpatients with CD4 count < 100 cells/mm3 (sensitivity: 40%; specificity: 87%).

     

    If LF-LAM test is positive: TB treatment should be initiated i Citation i. HIV-infected patients diagnosed with TB using the LF-LAM should be recorded as bacteriologically confirmed TB cases. .

    Due to the low sensitivity of the LF-LAM test, a negative result does not rule out TB. The test does not provide information on drug susceptibility. Therefore, all above-mentioned patients should be tested with an RMT, regardless of whether the LF-LAM result is positive or negative. 

    3.1.9 Medical imaging

    Radiography

    Chest x-ray (CXR) is used to:

    • Detect abnormalities suggestive of PTB and other intra-thoracic TB localisations (pleural, pericardial, miliary).
    • Evaluate the severity of intra-thoracic lesions.

    It is particularly useful in the diagnosis of PTB in children (Chapter 5).

     

    For PTB, CXR has a higher sensitivity than TB symptoms [23] Citation 23. World Health Organization. Chest radiography in tuberculosis detection – summary of current WHO recommendations and guidance on programmatic approaches. I. World Health Organization, 2016.
    https://apps.who.int/iris/bitstream/handle/10665/252424/9789241511506-eng.pdf;jsessionid=8552A4DE3F2B289132DA4342DBD962F9?sequence=1
    : a patient with a normal CXR is unlikely to have PTB. For this reason, it can also be used as a screening tool (Chapter 6) and a triaging tool to identify patients with respiratory symptoms eligible for an RMT.

     

    CXR is also used to:

    • Evaluate the response to TB treatment.
    • Look for possible complications in case of worsening respiratory symptoms (pneumothorax, tracheal stenosis, etc.).

     

    CXR has several limitations:

    • Low specificity: except for cavities or miliary TB, which are specific to TB other abnormalities seen on CXR may be due to other pulmonary diseases. 
    • Variable quality, depending on several factors:
      • equipment and supply
      • positioning (obtaining quality CXR in children is challenging)
      • reader training and proficiency
    • Difficulty distinguishing active from healed lesions
    • Error rate of approximately 20% [24] Citation 24. World Health Organization. Koppaka R, Bock N. How reliable is chest radiography? In: Frieden T, editor. Toman’s tuberculosis: case detection, treatment, and monitoring. Questions and answers, second edition. Geneva: World Health Organization; 2004. p. 51-60. https://apps.who.int/iris/bitstream/handle/10665/42701/9241546034.pdf?sequence=1 (specialists’ under/over-reading of the film)

     

    When available, digital CXR has advantages over x-ray films:

    • Consistent quality
    • Easier image archiving
    • No need for reagents and films
    • Rapid transmission for teleconsultation and specialist advice
    • Immediate results; possibility to screen large numbers of people within a short timeframe
    • Lower radiation exposure for staff and patients.

     

    Interpretation of digital CXR can be assisted by computer-aided detection (CAD) software packages. CAD analyses CXR for the presence of PTB-compatible abnormalities, and divides images into “normal” and “abnormal”, thereby reducing the number of CXR that need to be read by a clinician. CAD is as sensitive as a radiologist [25] Citation 25. World Health Organization. Rapid communication on systematic screening for tuberculosis. Geneva; 2020. https://www.who.int/publications/i/item/rapid-communication-on-the-systematic-screening-for-tuberculosis .

     

    Computer-aided CXR interpretation assists clinicians when all CXR cannot be read by a radiologist.

    However, a radiologist should be consulted locally or via telemedicine to interpret difficult CXR (e.g. in children).

     

    Bone x-ray is used to diagnose and evaluate severity of bone and/or joint TB and assess treatment response.

    Ultrasound

    Ultrasound (including point-of-care ultrasound, POCUS) may be useful in:

    • PTB: pulmonary consolidation can support the diagnosis of PTB.
    • EPTB: if suspected pleural/pericardial effusion or abdominal TB in children and adults, particularly in immunocompromised patients (e.g. HIV-infection, malnutrition).

     

    Table 3.8 – Medical imaging findings suggestive of TB

     

    Sites

    Findings

    Pulmonary TB

    Children

    See Chapter 5

    Adolescents and adults

    CXR can show:

    • Infiltrates typically located in apical and posterior segment of upper lobes and superior segments of lower lobes.
    • Cavities (specific for TB), patchy, poorly defined consolidations.

    Patients with TB/HIV

    As above.

    • In advanced immunodeficiency, infiltrates tend to be more homogeneous, diffuse and located in the lower lungs.
    • Less cavities than in non-HIV-infected patients.
    • Mediastinal and hilar lymphadenopathy may be observed.
    • Miliary pattern.

    Miliary TB

    CXR can show miliary nodules (1-3 mm in diameter) disseminated in both fields and uniformly distributed throughout the lung.

    Pleural effusion

    • CXR: effusion (even with minimal clinical signs):
      • Mostly unilateral.
      • Obliteration of costophrenic angle.
      • Opacity with curved upper margin.
    • Ultrasound: anechogenic fluid on the costophrenic angle (may be echogenic in empyema).

    Pericardial effusion

    • CXR: cardiac silhouette enlargement, “water bottle” silhouette (very large effusions).
    • Ultrasound: anechogenic fluid around the heart (may be echogenic if purulent).

    Bone/joint TB

    X-ray can show:

    • Any bone/joint: osteopenia (demineralization), bone destruction with relative preservation of cartilage space.
    • Spine: destruction of an inter-vertebral disk, osteopenia, irregularity of bone margin, bone destruction, paravertebral abscesses.

    Abdominal TB

    Ultrasound can show enlarged lymph nodes consistent with TB (and other diseases, especially in HIV infection), bowel wall thickening (ileo-caecal region), hypoechogenic micro-abscesses of liver and/or spleen, ascites.

     

    Notes:

    • Radiographical and ultrasound findings of EPTB are non-specific. A differential diagnosis should always be considered.  
    • In HIV-infected patients in settings of high TB prevalence, pleural/pericardial effusion, enlarged abdominal lymph nodes, splenic microabscesses, and ascites are highly suggestive of EPTB [26] Citation 26. Heller T, Wallrauch C, Goblirsch S, Brunetti E. Focused assessment with sonography for HIV-associated tuberculosis (FASH): a short protocol and a pictorial review. Crit Ultrasound J. 2012 Nov 21;4(1):21.
      https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3554543/pdf/2036-7902-4-21.pdf
      .
    • Adolescents typically have CXR abnormalities similar to those found in adults, however, they may also have abnormalities commonly seen in children, such as enlarged hilar lymph nodes.

    3.1.10 Other laboratory tests on tissues and body fluids

    The diagnosis of TB can be supported by biological tests performed on tissues or body fluids.

     

    Table 3.9 – Findings suggestive of TB in tissues or body fluids

     

    Tissues/fluids

    Findings

    Lymph node

    Cytology: granulomatous tissue, presence of giant Langhans cells, and/or caseous necrosis.
    AFBs are not always found by microscopy.
    See Lymph node fine needle aspiration, Appendix 7.

    CSF

    • Clear, hyper-concentrated liquid.
    • High protein level > 0.40 g/l (see Pandy test, Appendix 8).
    • Low glucose < 60 mg/l.
    • Ratio CSF glucose/blood glucose < 0.5.
    • Between 100 and 1,000 white cells/mm3, of which over 80% are lymphocytes.

    In HIV-infected patients, rule out cryptococcal meningitis.

    Peritoneal fluid

    • Translucent, yellow-coloured liquid.
    • Exudate rich in lymphocytes, usually > 300 white cells/mm3; Rivalta test positive (Appendix 8).
    • Serum-ascites albumin gradient (SAAG):

    < 1.1 g/dl: consistent with TB (and many other conditions).

    > 1.1 g/dl: peritoneal TB unlikely.

    • Adenosine deaminase (ADA) > 39 U/l, likely due to TB [27] Citation 27. Riquelme A, et al. Value of adenosine deaminase (ADA) in ascitic fluid for the diagnosis of tuberculous peritonitis: a meta-analysis. J Clin Gastroenterol. 2006 Sep; 40(8):705-10. .

    Pleural fluid

    • Straw-coloured fluid.
    • High protein level ≥ 30 g/l (Rivalta test, Appendix 8).
    • Rich in white cells (1,000-2,500/mm3), with predominant lymphocytes.
    • ADA typically > 50 U/l. Pleural effusion with an ADA < 40 U/l is much less likely due to TB. The specificity is increased when ADA is > 50 U/l and the lymphocyte-neutrophil ratio is > 0.75 [28] Citation 28. Porcel JM. Tuberculous pleural effusion. Lung. 2009 Sep-Oct;187(5):263-70. .

     

    Notes:

    • ADA levels increase in TB. ADA is therefore a surrogate marker for TB in pleural and peritoneal fluids. Although not widely available, kits can be purchased to perform the test if a spectrophotometer is available.
    • The sensitivity of ADA in peritoneal fluid is lower in patients with cirrhosis.
    • HIV-infected patients may have lower levels of ADA.

     

    Footnotes
    • (a)When microscopy is the only diagnostic test available, specimens should be sent to a facility with capacity to perform RMTs.
    • (b)For more information, see: Global Laboratory Initiative. Line probe assays for drug resistant tuberculosis detection Interpretation and reporting guide for laboratory staff and clinicians. http://stoptb.org/wg/gli/assets/documents/LPA_test_web_ready.pdf
    • (c)For more information :
      • WHO catalogue of mutations in M. tuberculosis complex and their association with drug resistance: https://www.who.int/publications/i/item/9789240028173
      • Relational Sequencing TB (ReSeqTB) Data Platform: https://c-path.org/programs/cptr/cptr-tools/databases/relational-sequencing-tb-data-platform-reseqtb/
    • (d)Microscopic observation of drug susceptibility (MODS), nitrate reductase assay (NRA), thin layer agar and colorimetric redox indicator (CRI).
    • (e)LAM antigen is a component of the mycobacterial cell walls released by M. tuberculosis then excreted by the kidneys.
    • (f)Alere Determine® TB LAM Ag (Alere Inc, Waltham, MA, USA).
    • (g)Seriously ill: respiratory rate > 30/minute, temperature > 39 °C, heart rate > 120/minute and unable to walk unaided.
    • (h)For children > 5 years and adults: CD4 count < 200 cells/mm3 or a WHO clinical stage 3 or 4. All children < 5 years are considered as having advanced HIV disease.
    • (i)HIV-infected patients diagnosed with TB using the LF-LAM should be recorded as bacteriologically confirmed TB cases.
    • (a)Mutations in other genes can result in resistance to thionamides. Consequently, absence of inhA mutation does not rule out resistance.
    • (b)Specific mutations in gyrA (e.g. mutations recognized by the probes MUT3B, 3C, 3D) are associated with high-level fluoroquinolones resistance.
      RMTs have a good specificity, but are less sensitive than culture. Their various levels of complexity determine their use at different levels of health facilities. Low complexity RMTs are preferred in routine practice.
    • (c) High- and low-level resistance detected by gDST and pDST.
    • (d) Rarely available in resource-limited settings.
    • (e)Most mutations conferring resistance to thionamides are not detected by RMTs.
    References