Tuberculosis remains a leading cause of mortality globally, even though effective drugs and preventive measures are available. However, an early and reliable diagnosis of tuberculosis is crucial for both appropriate treatment and effective disease control. Diagnosis of TB should include taking an anamnesis, a physical examination and a chest x-ray, as well as a tuberculin skin test. Microbiological examination of a sputum or other appropriate sample, various sero-diagnostic tests and newly-developed nucleic acid based tests facilitate diagnosis. In this article we review the various clinical laboratory tests available.
by Meghna Patel, Anita Joshi & Udaikumar Padigel


Microbiological tests
Microbiological techniques to facilitate diagnosis include the staining of smears from clinical samples (e.g. sputum, aspirated fluid etc.) for Acid Fast Bacilli (AFB) using the Ziehl – Neelsen (Z-N) method, and culture of the organism. This method has the advantages of simplicity, easy availability and speed. It is also the best method for monitoring drug response in DOTS (Directly Observed Treatment), where the pre-treatment  smear has revealed AFB [Figure 1].
An AFB smear report is not only of great value in diagnosis and prognosis, but also helps to grade the infection. The scale recommended by RNTCP (Revised National Tuberculosis Control Programme) is as follows:
3+  : > 10 AFB/ oil immersion field
2+  : 1-10 AFB/ oil immersion field
1+  : 10-99 AFB/ 100 oil immersion fields
Positive scanty: 1-9 AFB/ 100 oil immersion fields
Negative: no AFB/ 100 oil immersion fields

In order to detect 1-3 organisms in 300 oil immersion fields, the concentration of the organism must be 5,000-10,000/mL; this requires a trained and skilled technician. The test also has limitations for the diagnosis of paucibacillary tuberculosis and extra pulmonary tuberculosis, and cannot distinguish M. tuberculosis from other AFBs. The overall sensitivity of the smear technique is 40-60 %. It can be marginally increased by the use of fluorochrome stain, which allows M. tuberculosis to be easily visible at a comparatively lower magnification i.e. with a high power objective. However, the high cost of a suitable microscope and reagents, as well as the need for skilled personnel, can be restrictive. A negative smear result does not exclude the possibility of TB.

Bacterial culture
Culture of M. tuberculosis remains a gold standard for more sensitive detection, proper identification of Mycobacterium species and antibiotic sensitivity testing. Sputum, aspirated fluids and CSF are the main samples used for isolation of M. tuberculosis on conventional solid or liquid culture media [Figure 2]. The solid culture media available include LJ (Lowenstein – Jensen) medium and Middlebrook’s 7H10, 7H11 agar medium. Cultivation of M. tuberculosis on LJ medium is simple and economical but requires a prolonged incubation of 3-4 weeks, with an additional 3-4 weeks needed for antibiotic sensitivity testing. In addition, the success of the test largely depends on the decontamination technique, the reagents and the centrifugation method used for processing specimens before inoculating the culture medium. Middlebrook’s series of liquid media can decrease turnaround time somewhat but require a CO2 incubator and the prescribed antibiotics. Septi-check consists of specially designed bottles, enriched media and predetermined antibiotics, and allows differentiation of M. tuberculosis from MOTT (Mycobacterium other than tuberculosis), other respiratory pathogens and even contaminants.

Attempts to shorten the time taken to culture M. tuberculosis have resulted in development of automated systems such as the BACTEC, which detects radioactive CO2 generated by bacteria grown on a radio-labelled carbon source. The Mycobacterial Growth Indicator Tube (MGIT) 960 TB test employs a new fluorescent indicator based on oxygen quenching with fluorescent dye. The technique is much faster but needs monitoring every hour. This system, and similar radiometric- and fluorescence-based culture systems allow rapid isolation and antibiotic sensitivity testing, and differentiate M. tuberculosis from MOTT; automated and manual systems are available. However, appropriate infrastructure is essential, and high costs are incurred for instruments and consumables. A specialised disposal protocol for radioactive isotopes also restricts their use in routine pathology laboratories.

Tuberculin skin test
The Mantoux tuberculin test (MT) is the most commonly used skin test to facilitate TB diagnosis; it has replaced the former Heaf test and multiple puncture tests such as the Tine test. MT is a delayed hypersensitivity test. Tuberculin (a glycerol extract of tubercle bacilli) is injected intra-dermally in the forearm, and the subject’s immune response is observed and evaluated for the presence of sensitised T lymphocytes indicating past exposure to M. tuberculosis. Since it was first introduced, this test has undergone continuous refinement in the preparation, purification and formulation, standardisation and dosage of the test antigen, as well as the test’s interpretation and indication for use. The current test antigens suggested by WHO are purified protein derivatives PPD-RT-23 and PPD-S. All tuberculins have been standardised in “Tuberculin units” against the International Standard. This maintains uniformity and comparability of global test results. SPAN Diagnostics has been providing Tuberculin PPD in 1 TU, 2 TU, 5TU and 10 TU dosages for three decades. The WHO, however, has recommended a lower strength of tuberculin (1 TU) for standard MT in India because of the greater prevalence of atypical, as well as typical Mycobacteria infections in the population. This leads to high sensitivity but low specificity [4]. Two to three days after the intradermal tuberculin injection, the site should be carefully examined by a trained healthcare worker. In TB sensitised individuals the tuberculin injection produces a wheal and induration. The diameter of the indurated area should be measured across the forearm (perpendicular to the long axis). Erythema should not be measured. All reactions should be recorded in millimeters, even those classified as negative (0 mm). A positive reaction is indicated by an induration whose size will depend upon the dose of Tuberculin PPD injected and the prevalence of mycobacterial infections in the given population. The interpretation recommended by the Revised National Tuberculosis Control Programme (RNTCP) guidelines for India considers the size of induration as well as the probability of prior exposure to M. tuberculosis (immunological and BCG immunisation status).

An induration of 15 mm and above indicates infection with tubercle bacilli, irrespective of BCG vaccination status [Figure 3]. An induration of 10-14 mm could be attributable to cross-reactivity with other species of Mycobacteria, BCG-induced sensitivity or an infection with M. tuberculosis, especially if there has been contact with smear positive cases of pulmonary tuberculosis and an X-ray is consistent with active tuberculosis. An induration of 5-9 mm is mainly due to cross reactivity with other species of Mycobacterium and/or BCG vaccination, or infection with tubercle bacilli in subjects with immuno-suppression. No induration, or an induration of less than 5 mm, indicates absence of any type of mycobacterial infection except in children with a severe degree of immune-suppression. The MT is not indicated in individuals who had a previous positive reaction, patients with previous TB infection and infants under 12 weeks. Span’s tuberculin PPD is a ready-to-use solution for performing MT. The source material is calibrated against batch RT-23 manufactured by Statens Serum Institute, Denmark. It is diluted with a special buffer containing Tween 80 as a stabiliser.

Serological tests
Serodiagnostic techniques have been extensively investigated and are widely used for confirming TB infection. Serological diagnosis of TB is based on immunological detection of antibodies against M. tuberculosis using different assay systems. These include lateral flow or immunochromatography (ICT), Flow-through or Immunofiltration, Solid Phase Immunosorbent Assay (SPIA) or Dot Blot Assay, and ELISA. Various antigens are utilised in these assays.

With the advent of advanced protein purification techniques and recombinant DNA technology, previous problems caused by use of impure antigens in serological tests can be overcome. A-60, 38 kDa and many other similar recombinant protein antigens and lipoarabinomannan (LAM) are some of the most commonly used antigens in serodiagnostic tests for tuberculosis. A-60 is a complex mixture of antigens, which contains the main thermostable component of PPD. Although this is used in various commercial tests, it exhibits cross reactivity with antibodies against Nocardia and Corynebacterium spp [6, 7]. LAM is found in the cell wall of Mycobacteria. It can be separated with chromatography for use in serodiagnosis of tuberculosis. Although it is specific to the genus Mycobacterium, it cross reacts with many species other than M. tuberculosis, including M. leprae. 38 kDa is a species-specific antigen of M. tuberculosis. The gene of 38 kDa has been cloned and over expressed in E. coli. This recombinant product has been used successfully in many commercially available tests.

Some other recombinant protein antigens including antigen p90, antigen 19 kDa, antigen p32 and antigen 16 kDa have also been tested for their utility in immunodiagnosis of TB [8,9]. The search for useful antigens continues, as it has repeatedly been shown that single antigen assays are not completely satisfactory. Antigen A-60 based tests are still in use, but due to cross reactivity, a mixture of 38 kDa and LAM is considered as a better option. 38 kDA is specific to the Mycobacterium complex and LAM is specific to the genus Mycobacterium.

The major challenge for immunodiagnostic tests is to distinguish between M. tuberculosis and MOTT bacilli and to distinguish between active and latent tuberculosis. Reproducible methods for purification of Mycobacterial antigens are yet to be developed, hence the results of most currently available assays are variable depending on setting. Furthermore, in Mycobacterial disease the immune response generated appears to be associated with HLA Class II allotypes, and different patients recognise different antigens; the probability that all patients will recognise a single antigen is low. This is a major drawback for the development of antibody-based detection systems [10].

Rapid sero-diagnostic tests are of great value for screening purpose, especially in the field where quick and definitive diagnosis is required. First Generation Serological Tests for TB diagnosis made use of Lateral Flow ICT, Flow through ICT, and SPIA or Dot Blot Assays rather than classical ELISA. Lateral Flow ICT tests are based on chromatographic migration. As the test sample flows through the nitrocellulose membrane in the assembly of the device or dipstick after addition of the clearing buffer, the impregnated coloured antigen-colloidal gold signal reagent forms a complex with antibodies present in the sample. This antibody-antigen colloidal gold complex flows further to the test region where it is immobilised by purified or recombinant antigen coated on the test region of the membrane, forming a pink coloured band that indicates a positive result. In Flow through ICT, the selected antigens are immobilised on the nitrocellulose membrane, and form a complex with any antibodies in the sample. They then bind with the antigen colloidal gold signal reagent that is added in the next step; a positive result is visualised as a pink coloured dot. SPAN Diagnostic’s Crystal MTb and Signal MTb are rapid and extremely simple immuno chromatographic tests, which utilise a mixture of several highly purified recombinant protein antigens to achieve optimal sensitivity and specificity.

With SPIA or Dot Blot assays, the antigen is bound to the teeth of polystyrene combs. These combs are incubated in serum or whole blood samples to allow binding of the antigen and the anti-tubercular antibodies, if present, in the sample. The combs are then washed and incubated in a suspension of Colloidal Gold Signal Reagent. If antibodies to the antigen are present, a coloured spot is formed where the antigen is bound to the teeth of polystyrene combs. The intensity of the spot is then compared to the intensity of spot on a reference comb which is provided with the kit.
The Dot Blot Assay can be calibrated so that only samples from patients suffering from active tuberculosis produce a visible spot. This is an advantage of SPIA over lateral flow tests, as false positives due to past infections, latent infections and BCG vaccinations are minimised. Solid high density combs offer a maximum number of adsorption sites for antigen binding, which ensures optimal antibody capture and a stable platform for reaction since all the variables of flow based assays are eliminated. However the use of a single antigen and the pale coloured dot are the major disadvantages of this test format. Span’s Tb Spot version 3 is a simple and rapid SPIA test using mixture of five proprietary immunodominant recombinant antigens to detect antibodies to M. tuberculosis. The test is designed to avoid false positive results in a BCG-vaccinated population. It offers high specificity and sensitivity and detects pulmonary as well as extra pulmonary TB.

Indirect ELISA is a more sensitive technique for the detection of antibodies to M. tuberculosis. The most extensively used Mycobacterial antigen in this type of assay is A-60. In such ELISAs microwells coated with specific antigen are incubated with the sample, which forms antigen-antibody complexes if the sample contains the respective antibodies. After removal of non specific antibodies by a washing step, enzyme labelled anti-human IgG/IgM/IgA antibodies are added, which bind to the antigen-antibody complex. After a second washing step for removal of unbound enzyme labelled antibodies, a reagent that generates colour in presence of the enzyme is added. The colour is measured with an ELISA reader and is directly proportional to the concentration of the anti-tubercular antibodies present in the specimen. Even though they are more specific and sensitive, ELISA tests are not ideal for TB diagnosis, due to the long assay time and the difficulty of setting the positive cut-off point, which depends on the patient population and TB endemicity. Span’s Mycowell G, Mycowell A and Mycowell M are Tb ELSA kits, which determine the respective class of anti-tubercular antibodies. The test utilises a mixture of five highly purified proprietary recombinant protein antigens to achieve maximum sensitivity without compromising specificity. The test is specifically formulated to give no cross reactivity in a BCG vaccinated population, and to detect pulmonary as well as extra pulmonary TB. It can be performed as qualitative as well as quantitative test.
Second generation immunological tests for TB diagnosis are available in two test formats: SPIA and ELISA. The second generation SPIA is a rapid test detecting IgG antibodies against 38 kDa and LAM antigen. These tests are designed to detect levels of IgG only above the average local community titre. The test can be used for the quantitative estimation of IgG with the assistance of an interpretation chart provided with the kit. The second generation ELISA also uses a mixture of 38 kDa and LAM antigen. However the ELISA provides higher sensitivity and specificity, but is slower, more complex to perform and more expensive.
Serodiagnostic tests based on antigen determination have also proven useful, especially for epidemiological studies and for monitoring treatment [10,11,12]. Detection of antigens at a concentration of 3-20 ng/mL from different body fluids, such as CSF and pleural fluid, is possible. The most commonly used Mycobacterial antigens are extracted glycolipids, PPD, Ag5 (38kDa), A60, 45/47 kDa Ag, Ag Kp90 [8], 30kDa Ag [9], P32 Ag, cord factor (trehalose dimycolate) and LAM. These commonly targeted antigens are detected using sandwich ELISAs and latex agglutination reverse passive haemagglutination [12]. Recent reports of novel antigens include Rv2041c Ag [13], MPT 64 [14] and Rv3425. Span’s Crystal TB confirm is a rapid test, based on a lateral flow format, which detects MPT 63 and MPT 64 from liquid culture media inoculated with a sample containing M. tuberculosis, or from a colony collected from solid culture (like LJ medium) after emulsification in normal saline.

Nucleic acid based tests (NAT)
Advances in molecular biology have provided significant tools for direct detection, species identification and antibiotic sensitivity testing of TB. Molecular biological techniques for TB diagnosis include Amplification-based Nucleic Acid tests (NAT) and Non amplification-based nucleic acid tests using specific probes [15]. Initially only radio-labelled probes were available but now chemiluminescent probes are also available. Nucleic Acid Amplification tests are highly sensitive, specific (above 90%), rapid and widely available, and can be performed within one to two days. The tests involve amplification of specific gene regions using mycobacteria-specific primers following either multithermal or isothermal formats. PCR is widely used for amplification of the genetic target; it can amplify even a very small portion of a predetermined target region of M. tuberculosis-complex DNA. PCR based assays can rapidly detect as few as 1-10 organisms from sputum, broncho alveolar lavage, blood, cerebrospinal fluid, pleural fluid, and other fluid and tissue samples. Detection is usually carried out by either electrophoresing the amplified DNA products on agarose gel or by hybridisation using species-specific probes. Sequencing and Restriction fragment length polymorphism (RFLP) techniques are also used to analyse the amplified DNA products.

The PCR based assays usually target either DNA or rRNA. The most commonly used sequence for this purpose is IS6110 and 16S rRNA. PCR, followed by restriction enzyme analysis, that exploits the 65 kDa hsp and 16S rRNA genes for identification has also been used [9]. Several commercial and in-house developed multiplex assays are available, such as INNO-LiPA tests (Innogenetics) and Genotype-MTBC assays (Hain Life sciences), which can be used for identification of disease-causing mycobacterial species and their differentiation from MOTT. Modifications of conventional PCR, such as nested PCR and real time PCR, are also being used. Commercially available assays have proved to be reproducible and sensitive as well as accurate, and have already found worldwide acceptance. Because of the increasing use of NATs and the potential impact on patient care and public health, the US Centers for Disease Control and Prevention (CDC) recently recommended that NAT testing be performed on at least one respiratory specimen from each patient with signs and symptoms of pulmonary TB for whom a diagnosis of TB is being considered but has not yet been established, and for whom the result would alter case management or TB control activities, such as contactinvestigations [16].

The common objective of all technologies underlying in vitro amplification of Mycobacterial nucleic acids is to reduce the time necessary to detect the pathogen in clinical specimens, to increase specificity and sensitivity, and to simplify the test by automation. A positive direct amplified test in conjunction with an AFB-positive smear is highly predictive of TB. However, the Mycobacterial culture is further needed as a confirmatory test and for susceptibility testing. A negative NAT with a positive AFB-positive smear indicates that the AFB is probably another species of Mycobacterium / MOTT.

Applications that target RNA are expected to be more sensitive, because RNA already occurs in high copy numbers in bacterial cells. However, it is often the case that a higher analytical sensitivity does not necessarily improve clinical sensitivity. The limiting factor for all these techniques remains the same. An efficient sample processing is the prerequisite for high sensitivity and specificity of these tests. The loss of bacteria during sample processing, or the presence of inhibitors, strongly influence the sensitivity of NATs. Furthermore, these tests require experienced referral laboratories that have appropriate expertise, personnel, laboratory infrastructure and contamination control measures.

Thus TB diagnosis is not straightforward and is continuously being surrounded by uncertainty in spite of the availability of various tests. There are several sensitive and specific tests available, which can assist in diagnosis of tuberculosis. However, more than one test is required to confirm the disease. AFB staining must be combined with serodiagnostic tests and NATs to establish that an active tuberculosis infection is present, and for initiation of early treatment. Culture still remains the gold standard: every attempt should be made to isolate the organism fromthe specimen.

References
1. Banerjee S et al. Sero diagnosis of Tuberculosis using two ELISA systems. Indian J Clin Biochem 2003; 18: 48-53.
2. WHO. Tuberculosis facts 2009.
3. US Department of health, education and welfare, Food and Drug Administration. Skin test antigens: Proposed implementation of efficacy review. Fed Reg 1977; 42: 674-723.
4. Ten Dam HG. Surveillance of tuberculosis by means of tuberculin surveys. WHO/TB/85. 145, Geneva, WHO 1985.
5. Dooley T et al. Multidrug-resistant tuberculosis. Ann Intern Med 1992; 117: 257-  259.
6. Guilleron M, Usdin M, Arkinstall J. Tuberculosis diagnosis and drug sensitivity testing: an overview of the current diagnostic pipeline. In Campaign for access to essential medicines Oct 2006.
7. Mark D, Perkins MD. New diagnostic tools for tuberculosis. Int J Tuberc Lung Dis 2000; 4(12): s182-s188.
8. Arikan et al. Anti-Kp 90 Ig A antibodies in the diagnosis of active tuberculosis. Chest 1998;  114:1253-1257.
9. Bothamley I et al. Serodiagnostic value of the 19 kilodalton antigen of Mycobacterium tuberculosis in Indian patients. Eur J Clin Microbial Infect Dis 1992;
11:912-915.
10. Boehme C et al. Detection of Mycobacterial lipoarabinomannan with an antigencapture ELISA in unprocessed urine of Tanzanian patients with suspected tuberculosis. Trans R Soc Trop Med Hyg (2005) 99: 893-900.
11. Kadival GV et al. Diagnosis of tuberculosis using polyclonal and monoclonal antibodies. Trop Med Parasitol 1990; 41: 363-365.
12. Sada E et al. Detection of lipoarabinomannan as a diagnostic test for tuberculosis. J Clin Microbiol 1992; 30: 2415-2418.
13. Kim SY et al. Identification of Rv2041c, a novel immunogenic antigen from Mycobacterium tuberculosis with serodiagnostic potential. Scand J Immunol 2009; 70(5): 457-64.
14. Ismail NA et al. Use of an immunochromatographic kit for the rapid detection of Mycobacterium tuberculosis from broth cultures. Int J Tuberc Lung Dis 2009;  
13(8): 1045-7.
15. McFadden JJ, Kunze Z, Seechurn P. DNA probes for detection and identification. In: Molecular Biology of the Mycobacteria; J McFadden (Ed), Surrey University Press, UK, (1990) 139-172.
16. CDC report, 2008.

The authors
Meghna Patel
SPAN Education and Research Center
and
Anita Joshi & Udaikumar Padigel
SPAN diagnostics Ltd
Surat, Gujarat, India