Sputum processing prior to Mycobacterium tuberculosis detection by culture or nucleic acid amplification testing: a narrative review

Abstract

Sputum is a complex specimen consisting of a primary network comprised of linked mucin molecules and a secondary network composed of filamentous actin, cell debris and DNA. Other components of sputum are leukocytes, proteoglycans, inflammatory mediators and elastin fibres. Embedded within this matrix are bacteria which are targeted for clinical diagnosis. This is further complicated in tuberculosis as mycobacteria frequently clump within the specimen resulting in unequal distribution. Efficient release of bacteria from sputum specimens requires chemical and/or mechanical breakdown of both primary and secondary sputum networks. This review gives an overview of the composition of sputum and of various methods that have been used for digesting sputum prior to testing for tuberculosis.

Keywords

Sputum, Mucolytics, Peptide mucolytics

Introduction

Microscopy is currently the most widely used tool for the diagnosis of TB. Although lacking sensitivity, smear microscopy is inexpensive, relatively simple to perform and a rapid diagnostic test [1,2] that can be performed close to the point-of-care (POC). Culturing Mycobacterium tuberculosis (M. tuberculosis) is a much more sensitive means of diagnosis than sputum smear microscopy and is the reference standard for diagnosis of M. tuberculosis [3]. However, M. tuberculosis is a slow growing organism [2] and culture of clinical specimens typically takes from one to six weeks, depending on the bacterial load in the sample. Molecular methods (such as nucleic acid amplification (NAA) tests) have a rapid turnaround time and are thus attractive diagnostic tools. However, NAA tests often lack sensitivity. Theoretically, NAA tests require small quantities of bacteria [4,5] and until recently (with the introduction of the Xpert MTB/RIF test (Cepheid) [6] have not been widely adopted [7]. Optimal specimen processing protocols are key for the sensitive detection of M. tuberculosis bacteria by NAA testing, and require extraction, concentration and purification steps.

Specimen processing prior to culture and NAA testing for TB

Specimen processing is a critical and underestimated step when attempting to isolate M. tuberculosis from sputum [8,9]. Development of optimal sample processing is not sufficiently prioritised although there is an urgent need for it to go hand-in-hand with novel detection systems [9].

Although M. tuberculosis has the ability to infect a range of organs, the most common form of tuberculosis (TB) in adults is pulmonary TB [9], with sputum the specimen most frequently collected for diagnosis. Inefficient specimen processing will hamper the release of M. tuberculosis bacteria and or DNA from specimens, which will result in poor sensitivity of diagnostic tests. This is particularly relevant for patients with paucibacillary TB, which is common in children and in adults co-infected with HIV.

Theoretically, NAA tests require small quantities of bacteria [4,5] however, in practice they generally have suboptimal sensitivity in smear negative sputum specimens [10], which may, in part be due to the relative inefficiency of the sputum processing protocols used.

Sputum

Sputum is a mixture of mucus with other endogenous or exogenous components which may include transudated and exudated fluids, a range of local and migrated cells, microorganisms, necrotic tissues or cells, aspirated vomitus and other foreign particles [11].

Respiratory mucus forms part of the innate immune system [12]. It is divided into two phases viz, a sol (fluid) and a gel phase [13]. Respiratory mucus is made up of water, ions, proteins, lipids, enzymes, immunoglobulins and large glycoproteins of varying sizes [13-16]. These glycoproteins are called mucins.

Mucins form a dense, protective barrier preventing bacterial attachment to cells lining the respiratory tract [17]. Mucus traps bacteria and debris which are naturally moved out of the respiratory tract by ciliary clearance [12].

There are approximately 20 genes involved in human mucin expression. The glycoproteins encoded by the mucin genes are divided into three major families namely; membrane associated mucins (MUC1, MUC4, MUC11, MUC13, MUC15 and MUC20), non-gel forming mucins (MUC7) and gel-forming mucins (MUC5AC and MUC5B, MUC2, MUC8 and MUC19) [18]. The gel forming mucins, MUC5AC and MUC5B, comprise 79% of normal human respiratory secretions [19,20]. The MUC5AC gene is expressed in and limited to the goblet cells of the lung while MUC5B is expressed in the bronchiolar epithelium and submucosal glands [21].

In order for efficient ciliary clearance to occur, mucus must possess high elastic recoil together with low viscosity [22]. However, in the presence of disease (such as cystic fibrosis and bronchitis), alterations of the terminal glycosylations of mucins occur resulting in a physical change in the composition and nature of mucus [12].

Physiological or structural damage of cilia may occur during lung disease [23] and bacterial infections [24]. When cilia are damaged, ciliary clearance does not occur naturally and the mucus becomes trapped in the airways. In these cases cough may assist with clearance of mucus [23,25,26]. Further, during respiratory disease there may be an increase in the number of goblet cells which results in hypersecretion of mucus [27-29]. Mucin release can be regulated by irritant gases, inflammatory mediators, arachidonic acid metabolites, platelet activating factor, tumor necrosis factor, bacterial proteases, reactive oxygen species, nucleotides, neuronal control and mechanical strain (For review see Kim et. al.) [30]. Cigarette smoke also induces mucus hypersecretion [31]. Mucus builds up in the airways [27] forming a mucus plug [17]. Expulsion of this plug may require expectoration [18]. This expectorate is called sputum [18].

Sputum is a viscoelastic solid [32] made up of mucus [33], leukocytes, cellular debris, bacteria, filamentous actin (F-actin), proteoglycans, DNA, inflammatory mediators [16,17,34] and elastin fibres [35]. Each sputum specimen is unique hence the components present in one specimen may not be a representative of all sputum specimens. The type and extent of a disease (for example necrotising lung diseases such as tuberculosis or pneumonia) may affect the quantity of sputum constituents, such as elastin [36].

The length of actin filaments found in sputum has been shown to correlate with sputum cohesiveness [37,38] although filamentousactin (F-actin) may not be present in all sputa [39]. Other fibres may be derived from DNA and DNP (deoxyribonucleoprotein) which are released during leukocyte degeneration, to form fibres which contribute to sputum viscosity in purulent specimens [40,41]. These fibres are not present in non-purulent (mucoid) sputa.

Stagnant sputum present in the airways is conducive to bacterial multiplication [42] which in turn attracts leukocytes. The enzyme myeloperoxidase (MPO) is released from degenerating granulocytes in inflammatory states [41,43-45], and imparts a green colour to sputum [46]. Sputum colour is indicative of the degree of bacterial burden [47]. Purulence (green sputum) appears to be associated with increased bacterial load [47-50] whereas mucoid (white, cream) and clear sputum contains fewer bacteria [47,51]. However, this may not always be the case as Brusse-Keizer et al [52]. showed no significant association between sputum colour and bacterial load. In sputum microscopy more than 25 leukocytes per high power field is indicative of infection [53]. In patients undergoing therapy there may be reduced inflammation [54]; improved mucus clearance [54,55]; a decrease in purulence due to a reduced bacterial burden [45] as well as conversion from purulent to mucoid sputum [41].

Sputum Processing

Sputum processing prior to routine microbiological testing

Sputum processing in the diagnostic laboratory for non-tuberculosis specimens usually includes Gram stain and direct inoculation onto agar [56]. Sputum processing for mycobacterial culture however, involves liquefaction, decontamination, neutralisation and concentration. During liquefaction a mucolytic agent is added to sputum to release bacteria that may be trapped within the complex sputum network. This mucolytic is often also a decontaminant that kills contaminating microorganisms that might affect downstream mycobacterial culture. Petroff’s method which employs 4% sodium hydroxide (NaOH) was previously used for liquefaction and decontamination but this concentration of NaOH is extremely harsh on the tubercle bacilli [57]. It was shown that decreasing the concentration of NaOH and adding a digestant, N-Acetyl-L-Cysteine (NALC), increases the likelihood of isolating M. tuberculosis. Ratnam et al. showed NaOH at a 1.5% concentration to be sufficient to prevent bacterial overgrowth during culture for M. tuberculosis for most sputum samples [56,58]. Following liquefaction and decontamination, a neutralising substance (e.g., phosphate buffer (pH 6.8) when using an alkaline digest) [59] is added in a timely manner in order to stop the decontamination process, thereby reducing the impact of the mucolytic on the viability of mycobacteria. Neutralisation is typically followed by concentrating the released bacteria in a cell pellet. This is usually achieved by centrifugation.

Sputum processing prior to NAA testing

Sputum processing for NAA tests typically involves sputum liquefaction (often with use of a mucolytic agent); M. tuberculosis cell lysis (mechanical and/or chemical lysis may be utilised); followed by DNA purification in order to concentrate DNA and remove potential PCR inhibitors.

If both culture and NAA testing are required on a specimen, sputum processing prior to NAA testing may follow similar processes to those used prior to mycobacterial culture (in which case the pellet can be split between culture and NAA testing) [60]. Pathak et al. has shown long-term storage of sputa at -20°C subsequent to NALC treatment results in greater yield of M. tuberculosis DNA [61].

If culture is not required, as cell viability is not a prerequisite for NAA tests, more stringent processing can be employed in order to kill mycobacteria (as a biosafety precaution) and maximize release of most of the bacteria from the sputum. For example, sputum can be collected directly into media, which preserve nucleic acid; however, these have generally not been well-validated for M. tuberculosis. Examples include PrimeStore^® Molecular Transport Medium (PSMTM; Longhorn Vaccines & Diagnostics, San Antonio, TX, USA)[62], cetylpyridinium chloride (CPC) [63] and Universal sample processing solution (USP) [64]. Primestore has some mycobactericidal activity [65,66]. The Universal sample processing (USP) solution has not been used as transport medium; however, it contains 4-6 M guanidinium hydrochloride (which may aid in cell lysis), 50 mM Tris/Cl (which maintains the pH), 25 mM EDTA (a chelating agent), 0.5% Sarkosyl (anionic detergent that disrupts cell membranes) and 0.1 M β-mercaptoethanol (which may reduce cysteine residues). Solutions containing Guanidinium have been shown to preserve nucleic acids at room temperature for prolonged periods of time [67] hence the USP solution may be adapted for use as a transport medium. CPC can also be utilised as a transport medium [63] and is compatible with the Xpert MTB/RIF assay [68] whereas the USP solution has been shown to be compatible with DNA as well as RNA isolation [64].

Chemical and mechanical sputum liquefaction

Sputa can be liquefied chemically (by the use of mucolytic agents) or mechanically.

Chemical liquefaction of sputum

Mucolytic agents are used to liquefy/digest sputum. Some mucolytics are administered orally (oral alpha-chymotrypsin) to improve mucociliary clearance in diseased individuals. Others are employed in vitro (NALC-NaOH) for microbiological investigations. Liquefaction results in a change in the biophysical properties of sputum usually by reduction of mucin molecules, fibrin, F-actin and DNA [69]. For mucin, this involves the separation of the intermolecular hydrogen bonds which link mucin molecules, which in turn results in a reduction of entanglement points and hence contributes to a decrease in viscosity [70].

There are different types of mucolytics, namely, classical and peptide mucolytics. Classical mucolytics act on the primary network, by digesting bonds linking the mucin network. Peptide mucolytics act on the secondary network, which comprises cellular debris, F-actin and DNA [69].

Heterogeneity within and between individual sputum samples is due to the intricate structure of mucus and the variable nature of the underlying pathology [71]. Therefore mucolytic agents may not be effective on all sputum samples as they may act on targets not present in a specific sputum specimen or as physical or chemical barriers may prevent the agent from accessing its target. As an example, Deoxyribonuclease (DNase) does not act on the mucoprotein gel, which may be present in certain sputum samples, and may therefore not be an efficient mucolytic for all sputum samples [41].

When samples are being processed for culture, the liquefaction process usually occurs concurrently with sputum decontamination. The liquefaction and/or decontaminating agent is typically added to the sputum specimen followed by incubation at room temperature for 15 - 20 minutes. Some investigators have suggested incubating viscous specimens at 37°C [72,73]. It is important to note that some liquefaction and decontamination procedures (Zephiran-Trisodium Phosphate, Sodium Lauryl Sulphate, Cetylpyridinium chloride or other quaternary ammonium compounds) are only compatible with egg-based culture media and cannot be used in conjunction with the Mycobacteria Growth Indicator Tube (MGIT™) system [59].

Currently the most widely used method for liquefaction and decontamination of sputum is that described by Kent and Kubica (1985) [74]. This method involves the use of NALC (0.5% final conc.) and NaOH (2% final conc.) together with sodium citrate (1.45% final conc.) [59].

The extent to which sputum liquefaction occurs can be measured by means of a viscometer. Examples of viscometers used in sputum liquefaction studies include the Consisto-viscometer [75]; the Brookfield viscometer [76] and the sputum consistometer [77]. Another means of determining sputum liquefaction is to take note of the liquid portion and the pellet following centrifugation. Smaller, more compact pellets with an increase in the fluid component may signify more efficient liquefaction.

Types of mucolytics

A summary of mucolytics is presented in Table 1. Even though a variety of liquefaction agents exists, NALC/NaOH and dithiothreitol (DTT) are the mucolytics that are most widely used for sputum liquefaction prior to M. tuberculosis culture and NAA testing. Although potent, these mucolytics are classical mucolytic agents which act on mucin and not on the secondary network present in sputum which contributes to sputum viscosity. Efficient reduction of the primary and secondary network present in sputum may theoretically result in greater sensitivity of M. tuberculosis culture and NAA tests.

Table 1. Mucolytic agents.

Classical mucolytics

Classical mucolytic agents, NALC, DTT and Thioredoxin (Trx), break the disulphide bonds that link the mucin monomers (Table 1). In vitro studies show NALC to be a potent mucolytic agent when used in conjunction with NaOH [57]. Although NALC is currently the most widely used mucolytic agent in sputum processing, there is contradictory evidence on its utility. A study by Lorian and Lacasse demonstrated that 0.5% NALC plus 2% NaOH liquefies sputa similarly to 2% NaOH [95]. However, Kubica et al. found that the addition of NALC to NaOH enhances liquefaction. Dippy and Davis showed a significant reduction in sputum viscosity with the use of 20% NALC (without the addition of NaOH). Digestion and decontamination described by Kent and Kubica includes the addition of 0.2% NALC/4% NaOH mucolytic to the sputum specimen at a 1:1 ratio. The specimen/mucolytic suspension is mixed (by inverting the tube several times or briefly vortexing) and incubated at room temperature for 15 minutes. Subsequent to incubation, a neutralising agent (sterile distilled water or phosphate buffer (pH 6.8)) is added to stop the action of NaOH on the tubercle bacilli. The suspension is thereafter centrifuged at 3000 xg for 15 minutes. The supernatant is removed and discarded safely and appropriately and the sediment is resuspended in phosphate buffer (pH 6.8).

In its native form NALC is a white crystalline powder which has a shelf life of up to 3 years if refrigerated. However a major disadvantage of NALC is that once solubilised, mucolytic activity is lost after 24 hours [59,74,96] resulting in the need for daily reconstitution of NALC/NaOH solution [97]. In addition, high concentrations of NALC and NaOH may affect detection of growing M. tuberculosis in the MGIT™ liquid culture detection system [59].

Sputum liquefaction using DTT has been shown to be more effective than NALC75, 79 even at low concentrations (0.1 M DTT vs. 1.2 M NALC) [75]. DTT has the ability to effectively liquefy sputum, showing a >90% reduction in sputum elasticity [32]. Another study which compared sputum processing prior to cytomorphological examination for diagnosing lung cancer, found sputum processing with DTT (0.2% final concentration) to be superior to NALC (1% final concentration), with regards to cellularity (P < 0.0001) [98].

Sputa from cystic fibrosis patients are thick and viscous and difficult to homogenise solely by chemical means [99]. In vitro studies have demonstrated that Thioredoxin (Trx) is able to liquefy thick, purulent sputa from cystic fibrosis patients more efficiently than NALC and DTT [79]. Liquefaction was noted using Trx concentrations as low as 1 μM with maximum efficacy at 30 μM. Both DTT and Trx have greater muco-active capacities than NALC. The authors suggested that this may be due to Trx and DTT being dithiols (having 2 redox-active cysteine residues) whereas NALC is a monothiol [79].

Peptide mucolytics

Peptide mucolytics act on the secondary network present in sputum. The secondary network is composed of cellular debris, F-actin and/or DNA. Various peptide mucolytics have been reported and are listed in Table 1.

DNase liquefies purulent sputum by digesting DNA/DNP fibres that contribute to sputum viscosity [40]. These fibres are formed by extracellular DNA/DNP from cells undergoing degeneration. The DNA present in intact cells is not affected by DNase digestion. However, when processing sputa for NAA tests, caution should be taken when employing DNAse, as cell free DNA of the target bacteria under investigation may be present in the specimen. DNase does not affect the viscosity of mucoid specimens as the enzyme has no effect on the mucoprotein gel [41].

The current literature on the mechanism of action of sodium hypochlorite (NaOCl, household bleach) on sputum is inconclusive. However, studies have demonstrated a reduction in debris visualised on smear microscopy when using NaOCl [84,85] as well as associated increased sensitivities when compared with direct smears [84,85,100]. This may be an indication that NaOCl acts as a peptide mucolytic, breaking down the secondary network formed in sputum. NaOCl has a dual function in sputum processing; it can be used as a digestant and a decontaminant (having the ability to kill contaminating bacteria as well as M. tuberculosis) [85]. A 2.5% (final concentration) NaOCl solution is able to decontaminate and digest purulent sputa within 30 minutes [85]. Although 2.5% NaOCl kills bacteria, including M. tuberculosis, M. tuberculosis does not lose its “acid-fastness”. However, direct treatment of slides with 5% NaOCl substantially reduced the number of bacilli visualized on smear microscopy [101]. This is probably due to exposure of a thin layer of specimen to 5% NaOCl as compared to processing an entire specimen.

Unclassified mucolytics

There are several mucolytics that have not been classified as classical nor peptide mucolytics (Table 1). The mechanisms of action of these mucolytics are mostly unknown. However, the Modified Jungmann’s method (ferrous sulphate, sulphuric acid and hydrogen peroxide) aids in breaking down cellular debris in the sputum and iodated compounds are said to, together with a protease (such as trypsin), change the nature of sputum proteins by inducing proteolysis. Data on the use of these mucolytics are scarce.

Commercial/proprietary mucolytics

The Xpert MTB/RIF sample reagent contains NaOH and isopropanol (concentrations are not known, proprietary information). Sputa collected from patients are treated with sample reagent for 15 minutes. The Xpert MTB/RIF sample reagent efficiently digests sputum and efficiently reduces viability of M. tuberculosis by 8 log [94].

Mechanical liquefaction of sputa

Mechanical digestion is another means of liquefying sputum and is often used in conjunction with chemical digestion. Sputa can be liquefied by vortexing with the aid of glass beads or by ultrasonication. This can be done in the presence or absence of a mucolytic agent.

The tendency of mycobacteria to clump within specimens might affect sensitivity when samples are split for routine microbiological investigations. In one study, sputa were homogenised by vortexing with NALC alone, NALC and glass beads or glass beads alone [102]. Samples were thereafter split in two, serially diluted and cultured onto Middlebrook 7H11 agar [102]. The authors concluded that chemical and or mechanical processing of sputa is equally effective at recovering viable tubercle bacilli and that there was no significant difference between the three methods [102].

Another study investigated the use of glass beads with and without DTT for the recovery of bacteria (not M. tuberculosis) from sputum. Glass beads alone were not as efficient for recovery of Haemophilius influenzae as glass beads with DTT treatment (3.8 Í 10⁸ CFU/ml vs 5.2 Í 10⁸ CFU/ml) [103]. H. influenzae has been previously used as a model for a fragile bacterial cell [104], yet recovery rates were good [103], indicating that fragile cells are able to withstand the harsh conditions of vortexing in the presence of glass beads. In contrast, the cell wall of M. tuberculosis is difficult to lyse (due to its high lipid content) and so likely to withstand this method.

Another method commonly used for mechanical digestion is sonication with the aid of a waterbath. An in-depth description of the mechanism of action of sonication on sputa is described in Baxter et al. Sputa may be sonicated in the presence or absence of a mucolytic agent. Nauwelaers et al. used Ultrasone waves (sonication) to liquefy sputa for extraction of Human Respiratory Syncytial Virus (RSV) RNA. Here DTT and PBS were added to sputa before being sonicated in an Adaptive Focused Acoustics (AFA™) instrument [105].

Although both NALC and DTT are potent mucolytics, they do not digest cystic fibrosis sputum efficiently [99]. However, complete homogenisation was noted when sputa treated with DTT were sonicated intermittently for 120 seconds (30 second intervals) [99]. These fully homogenised samples also showed a reduction in Cq values (mean: 4.25 cycles) by a real-time PCR assay targeting Aspergillus species [99], indicating that mechanical lysis coupled with chemical lysis may improve the sensitivity of NAA tests.

The effect of filtration on sputa

Stepwise filtration of DTT-liquefied sputa with 40, 20 and 11 μm nylon net filters showed a reduction in squamous cells (present in saliva, 30-60 μm in size) but did not affect counts of cells of bronchial origin (approximately 10 μm in size) or differential cell counts [73]. Such filters are therefore likely to allow microorganisms to filter through, provided the microbial cells are not adherent to large cell debris. Filtration may aid microbiological tests by releasing microorganisms from sputa for subsequent culture, microscopy and NAA testing. A filtration system is also integrated into GeneXpert^® cartridges. This filter separates the M. tuberculosis bacilli from the rest of the liquefied sputum and its components thereby concentrating the bacilli [106] and eliminating the need for centrifugation.

The impact of sputum liquefaction and extraction protocols on the performance of NAA tests

The methods used to liquefy sputum, extract and purify DNA are likely to have a significant impact on downstream NAA testing. However, variations in study population, sample preparation, sample splitting, DNA target and PCR protocols make it very difficult to compare results on the utility of different methods used for sputum liquefaction and DNA extraction.

Although NALC is generally used for sputum processing prior to NAA tests, DTT has also been shown to be “PCR friendly”. Higher concentrations of NALC (> 0.5 g/L) and DTT (0.1 g/L) have been shown to result in PCR inhibition [107]. Two studies utilising DTT prior to NAA testing for M. tuberculosis, targeting the SecA gene and IS6110 element, yielded sensitivities greater than 95% with culture as a reference standard [108,109]. However, when NALC was used in a multi-site study involving identically spiked sputum, saliva and water specimens, participating laboratories yielded inconsistent positive results (ranging between 2% and 90%) for 103 Mycobacterium bovis BCG organisms [110]. Each laboratory employed their own sample preparation, DNA extraction, and PCR techniques but made use of the insertion sequence, IS6110, as the PCR target, upon request of the investigators. Table 2 provides a summary of four studies that utilised NALC and/or DTT as a mucolytic prior to nucleic acid amplification testing.

Table 2. A summary of sputum processing methods, and the resultant sensitivities and specificities of PCR methods employed in each of the studies.

Xiang et al. noted a significant reduction in RNA concentration when using 0.1% DTT for sputum processing as compared to specimens not treated with DTT, indicating that even low concentrations of DTT may affect the extraction of RNA [114]. Desjardin et al. showed that NALC/NaOH affects the recovery of M. tuberculosis mRNA but not rRNA. DNA yield was unaffected by the NALC/NaOH treatment [115]. The use of RNA as a target is attractive as there may be multiple copy numbers and RNA can be used to differentiate between dead and live organisms. This may be useful for identifying response to therapy and in patients with a recent history of TB treatment.

Paramagnetic particles (PMP) technology is another means of enriching M. tuberculosis cells and / or M. tuberculosis DNA from processed sputum specimens. PMP’s can be used to concentrate cells and /or DNA instead of centrifugation. TB-beads (Microsens Medtech Ltd, London, UK) used to concentrate M. tuberculosis bacilli from NALC-NaOH processed sputa for smearmicroscopy proved to be slightly inferior to conventional concentration by centrifugation (89.4% vs. 91.8% sensitivity) [116], but significantly improved the sensitivity of smear microscopy (P=0.002 and P<0.001) [117,118] when compared with microscopy of unconcentrated sputum. Ghodbane and Drancourt used the same TB-beads for concentration prior to culture and reported comparable results to centrifugation. There are no published results on the use of TB-beads prior to NAA testing [119].

Sputum processing at the point of care (POC)

Resource-poor countries require POC molecular diagnostics that are highly sensitive and specific (thus eliminating the need for culture), inexpensive, robust and simple to perform. These diagnostic tests should be performed at or close to the site of sputum collection and results should be given on the same day.

Mucolytic agents used at POC should be stable as well as safe. In addition, the mucolytic should be able to withstand extreme temperatures. The Xpert MTB/RIF assay reagents are stable between 2°C and 28°C hence its application as a POC test in countries with extreme temperatures is limited to those settings where reagents can be maintained at controlled temperatures.

Another important issue that needs to be taken into consideration for POC testing is biosafety and the use of hazardous chemicals. Patients as well as personnel performing diagnostic tests need to be protected from potentially infectious specimens and harmful reagents or chemicals that might be included in the POC test. Biosafety cabinets are impractical for use in low-income countries as they are costly and require regular maintenance. An example of appropriate processing reagents is the Xpert MTB/ RIF sample reagent buffer which is safe to handle (although high concentration of isopropanol requires dedicated shipping due to the flammable risk) and demonstrates 8 log killing efficiency [94]. Sample preparation is quick and easy and can be performed without the use of a centrifuge.

Many current sputum processing methods for diagnosing TB involve the use of a centrifuge to concentrate M. tuberculosis bacilli. Resource-poor settings require alternative means of concentrating M. tuberculosis bacilli for true POC testing. Alternative methods may include sedimentation [87,120,121], filtration or paramagnetic particle technology, which uses simple magnets to capture M. tuberculosis from liquefied sputa [116].

Conclusion

The sensitivity of sputum-based diagnostic tests for M. tuberculosis is largely dependent on the efficiency of sputum processing protocols. Efficient processing of samples leads to release of bacteria trapped within the complex sputum matrix. Efficient mucolytics include NALC and DTT. Chemical and / or mechanical digestion can be used; however a combination of both is more likely to result in enhanced homogenisation of sputum. Concentration of M. tuberculosis from sputum specimens for subsequent culture or NAA testing can be achieved by centrifugation, filtration or paramagnetic particles.

Acknowledgements

The authors would like to thank Prof Anwar Mall, Mrs Vanessa January, Ms Fadheela Patel and Mrs Charmaine Barthus for their valuable comments.

References

1. ATS (American Thoracic Society). Diagnostic Standards and Classification of Tuberculosis in Adults and Children.Am J Critical Care Med. 2000; 161:1376-1395.
2. FriedenTR, et al.Tuberculosis.Lancet 2003; 362:887-899.
3. GieRP, et al.The Challenge of Diagnosing Tuberculosis in Children: A Perspective from a High Incidence Area.PaediatrRespir Rev 2004;5:S147-S149.
4. Pai M andLing DI.Rapid Diagnosis of Extrapulmonary Tuberculosis using Nucleic Acid Amplification Tests: What is the Evidence?Future Microbiology 2008; 3:1-4.
5. Shinnick TM and Good RC.Diagnostic Mycobacteriology Laboratory Practices.Clinical Infectious Diseases 1995; 21:291-299.
6. Boehme CC, et al. Rapid Molecular Detection of Tuberculosis and Rifampin Resistance.The New England Journal of Medicine 2010; 363:1005-1015.
7. GrayJW. Childhood Tuberculosis and its Early Diagnosis.Clinical Biochemistry 2004; 37:450-455.
8. Niemz A, et al.Point-of-Care Nucleic Acid Testing for Infectious Diseases.Trends in Biotechnology 2011; 29:240-250.
9. Young DB, et al. Confronting the Scientific Obstacles to Global Control of Tuberculosis.The Journal of Clinical Investigation 2008; 118:1255-1265.
10. Tsara V, et al. Problems in Diagnosis and Treatment of Tuberculosis Infection.Hippokratia 2009; 13:20-22.
11. Farzan S.Cough and Sputum Production.In:H.K. Walker, W.D. Hall and J.W. Hurt (eds.) Clinical Methods: The History, Physical, and Laboratory Examinations. 1990;(3rd edn).
12. Rose MC and VoynowJA.Respiratory Tract Mucin Genes and Mucin Glycoproteins in Health and Disease.Physiological Reviews 2006; 86:245-278.
13. Lopez-vidrieroMT. Airway Mucus; Production and Composition.Chest 1981;80:799-804.
14. Gudis DA and Cohen NA.Cilia Dysfunction.Otolaryngologic Clinics of North America 2010; 43:461-472.
15. LewisRW.Lipid Composition of Human Bronchial Mucus.Lipids 1971; 6:859-861.
16. VoynowJA, et al. Regulation of Mucin Genes in Chronic Inflammatory Airway Diseases.Am J Respir Cell Mol Biol2006;34:661-665.
17. Henke MO, et al.MUC5AC and MUC5B Mucinsare decreased in Cystic Fibrosis Airway Secretions.Am J Respir Cell Mol Biol2004;31:86-91.
18. VoynowJAand Rubin BK.Mucins, Mucus, and Sputum.Chest 2009;135: 505-512.
19. Thornton DJ, et al. Respiratory Mucins: Identification of Core Proteins and Glycoforms.Biochemical Journal 1996;316:967-975.
20. Thornton DJ, et al. Identification of Two Glycoforms of the MUC5B Mucin in Respiratory Mucus.The Journal of Biological Chemistry 1997; 272:9561-9566.
21. Reid CJ, et al.Developmental Expression of Mucin Genes in the Human Respiratory Tract.Am J Respir Cell Mol Biol1997;17:592-598.
22. DulfanoMJ and Adler KB.Physical Properties of Sputum. VII. Rheologic Properties and Mucociliary Transport.The American Review of Respiratory Disease 1975; 112:341-347.
23. Mossberg B and Camner P. Impaired Mucociliary Transport As a Pathogenic Factor In Obstructive Pulmonary Diseases.Chest 1980; 77:265-266.
24. Wilson R, et al. Effect of Bacterial Products on Human Ciliary Function in vitro.Thorax 1985; 40:125-131.
25. AfzeliusBA and Mossberg B.Immotile Cilia.Thorax 1980; 35:401-404.
26. Stannard W and O’Callaghan C. Ciliary Function and the Role of Cilia in Clearance.Journal of Aerosol Medicine 2006; 19:110-115.
27. Aikawa T, et al.Marked Goblet Cell Hyperplasia with Mucus Accumulation in the Airways of Patients Who Died of Severe Acute Asthma Attack.Chest 1992; 101:916-921.
28. OrdoñezCL, et al. Mild and Moderate Asthma Is Associated with Airway Goblet Cell Hyperplasia and Abnormalities in Mucin Gene Expression.Am J RespirCrit Care Med2001;163:517-523.
29. Villenave R, et al. In vitro Modeling of Respiratory Syncytial Virus Infection of Pediatric Bronchial Epithelium, The Primary Target of Infection In vivo.PNAS 2012; 109:5040-5045.
30. Kim KC, et al. Airway Goblet Mucin: Its Structure and Regulation of Secretion.European Respiratory Journal 1997; 10:2644-2649.
31. Montalbano AM, et al.Cigarette Smoke Alters Non-Neuronal Cholinergic System Components Inducing MUC5AC Production in the H292 Cell Line.European Journal of Pharmacology 2014; 736:35–43.
32. Nielsen H, et al. Elastic Contributions Dominate the Viscoelastic Properties of Sputum from Cystic Fibrosis Patients.Biophysical Chemistry 2004; 112:193-200.
33. Murray PR and Washington JA.Microscopic and Baceriologic Analysis of Expectorated Sputum.Mayo Clinic Proceedings 1975; 50:339-344.
34. King M and Rubin BK.Pharmacological Approaches to Discovery and Development of New Mucolytic Agents.Advanced Drug Delivery Reviews 2002; 54: 1475-1490.
35. ShlaesDM, et al. Sputum Elastin Fibers and the Diagnosis of Necrotizing Pneumonia.Chest 1984; 85:763-766.
36. ShlaesDM, et al. Elastin Fibers in the Sputum of Patients with Necrotizing Pneumonia.Chest 1983; 83:885-889.
37. Kater A, et al.The Role of DNA and Actin Polymers on the Polymer Structure and Rheology of Cystic Fibrosis Sputum and Depolymerization by Gelsolin or Thymosin Beta 4.Annals of the New York Academy of Sciences 2007; 1112:140-153.
38. RubinBK, et al. Thymosin ß4 Sequesters Actin in Cystic Fibrosis Sputum and Decreases Sputum Cohesivity in Vitro. Chest 2006; 130:1433-1440.
39. TangJX, et al. Anionic Poly(amino acids)s Dissolve F-actin and DNA Bundles, Enhance DNase Activity, and Reduce the Viscosity of Cystic Fibrosis Sputum.American Journal of Physiology Lung Cellular and Molecular Physiology 2005; 289:L599-L605.
40. ArmstrongJBand WhiteJC. Liquefaction of Viscous Purulent Exudates by Deoxyribonuclease.Lancet 1950; 9:739-742.
41. Elmes PC and WhiteJC. Deoxyribonuclease in the Treatment of Purulent Bronchitis.Thorax 1953; 8:295-300.
42. FahyJV and Dickeybf.Airway Mucus Function and Dysfunction.New England Journal of Medicine 2010; 363: 2233-2247.
43. KlebanoffSJ. Myeloperoxidase: Friend and Foe.Journal of Leukocyte Biology 2005; 77:598-625.
44. Bresser P, et al.Airway Inflammation in Nonobstructive and Obstructive Chronic Bronchitis with Chronic Haemophilusinfluenzae Airway Infection Comparison with Noninfected Patients with Chronic Obstructive Pulmonary Disease.Am J RespirCrit Care Med2000; 162:947-952.
45. WhiteAJ, et al. Resolution of Bronchial Inflammation is Related to Bacterial Eradication Following Treatment of Exacerbations of Chronic Bronchitis.Thorax 2003;58:680-685.
46. RobertsonAJ.Green Sputum.Lancet 1952; 1:12-15.
47. Stockley RA, et al. Relationship of Sputum Color to Nature and Outpatient Management of Acute Exacerbations of COPD.Chest 2000; 117: 1638-1645.
48. Allegra L, et al.Sputum Colour as a Marker of Acute Bacterial Exacerbations of Chronic Obstructive Pulmonary Disease.Respiratory Medicine 2005; 99:742-747.
49. Currie DC, etal.Impaired Tracheobronchial Clearance in Bronchiectasis.Thorax 1987; 42:126-130.
50. Murray MP, et al. Sputum Colour: A Useful Clinical Tool in Non-cystic Fibrosis Bronchiectasis.European Respiratory Journal2009; 34:361-364.
51. Johnson AL, et al. Sputum Color: Potential Implications for Clinical Practice.Respiratory Care 2008; 53: 450-454.
52. Brusse-keizerMGJ, et al.Relation of Sputum Colour to Bacterial Load in Acute Exacerbations of COPD.Respiratory Medicine 2009; 103:601-606.
53. Joyce SM.Sputum Analysis and Culture.Annals of Emergency Medicine 1986;15:325-328.
54. Pereira AC, et al. Analysis of the Sputum and Inflammatory Alterations of the Airways in Patients with Common Variable Immunodeficiency and Bronchiectasis.Clinics 2009; 64:1155-1160.
55. Tamaoki J, et al.Effect of Clarithromycin on Sputum Production and its Rheological Properties in Chronic Respiratory Tract Infections.Antimicrobial Agents and Chemotherapy1995; 39:1688-1690.
56. Vandepitte J, et al.Lower Respiratory Tract Infections, Processing of Sputum in the Laboratory (for Non-Tuberculous Infections).In: Basic laboratory procedures in clinical bacteriology. WHO 2003;
57. Kubica GP, et al.Sputum Digestion and Decontamination with N-Acetyl-L-Cysteine-Sodium Hydroxide for Culture of Mycobacteria.The American Review of Respiratory Disease 1963; 87:775-779.
58. Ratnam S, et al. Simplified Acetylcysteine-Alkali Digestion-Decontamination Procedure for Isolation of Mycobacteria from Clinical Specimens.J ClinMicrobiol1987; 25:1428-1432.
59. SiddiqiSHand rüsch-gerdesS.FIND MGITTM Procedure Manual, Section V – Appendix B-4: Reagents for digestion decontamination. 2006;
60. Clarridge III JE, et al. Large-Scale Use of Polymerase Chain Reaction for Detection of Mycobacterium tuberculosis in a Routine Mycobacteriology Laboratory.J ClinMicrobiol1993; 31:2049-2056.
61. Pathak D, et al.Lysis of Tubercle Bacilli in Fresh and Stored Sputum Specimens: Implications for Diagnosing Tuberculosis in Stored and Paucibacillary Specimens by PCR.BMC 2007; 7:83-.
62. Daum LT, et al.Molecular Detection of Mycobacterium tuberculosis from Sputum Transported in PrimeStore® from Rural Settings.The International Journal of Tuberculosis and Lung Disease2015;19:552-557.
63. Pal N, et al.Transport and Storage of Sputum by using Cetylpyridinium Chloride for Isolation of Mycobacteria.Indian J PatholMicrobiol2009;52:59-61.
64. Chakravorty S and TyagiJS.Novel Use of GuanidiniumIsothiocyanate in the Isolation of Mycobacterium tuberculosis DNA from Clinical Material.FEMS Microbiology Letters 2001; 205:113-117.
65. Daum LT, et al.A Molecular Transprt Medium for Collection, Inactivation, Transport and Detection of Mycobacterium tuberculosis.International Journal of Tuberculosis and Lung Disease2014; 18:847-849.
66. OmarSV, et al. Laboratory Evaluation of a Specimen Transport Medium for Downstream Molecular Processing of Sputum Samples to Detect Mycobacterium tuberculosis.Journal of Microbiological Methods 2015; 117:57-63.
67. TuttleRM, et al.Preservation of Nucleic Acids for Polymerase Chain Reaction after Prolonged Storage at Room Temperature.Diagnostic Molecular Pathology 1998;7:302-309.
68. USAID TB Care II. The PIH Guide to the Medical Management of Multidrug-Resistant Tuberculosis, 2nd Edition. Partners in Health. Boston, USA. 2013;
69. Rubin BK.Mucolytics, Expectorants and Mucokinetic Medications.Respiratory Care 2007; 52:859-865.
70. Marriott C andRichardsJH.The Effects of Storage and of Potassium Iodide, Urea, N-Acetyl-Cysteine and Triton X-100 on the Viscosity of Bronchial Mucus.British Journal of Diseases of the Chest 1974; 68:171-182.
71. Broughton-Head VJ, et al. Unfractionated Heparin Reduces the Elasticity of Sputum from Patients with Cystic Fibrosis.American Journal of Physiology Lung Cellular and Molecular Physiology 2007; 293:L1240-L1249.
72. Haldar S, et al. Simplified Detection of Mycobacterium tuberculosis in Sputum Using Smear Microscopy and PCR with Molecular Beacons.Journal of Medical Microbiology 2007; 56:1356-1362.
73. Ronchi MC, et al.Sputum Processing: A New Method to Improve Cytospin Quality.Clinical and Experimental Allergy 2002; 32:674-680.
74. Kent PT and Kubica GP.A Guide for the Level III Laboratory. In: Public Health Mycobacteriology. Atlanta, GA: Centers for Disease Control, U.S. Department of Health and Human Services. 1985;
75. Hirsch SR, et al. Sputum Liquefying Agents: A Comparative in vitro Evaluation.The Journal of Laboratory and Clinical Medicine 1969; 74: 346-353.
76. Hurst GA, et al.Laboratory and Clinical Evaluation of the Mucolytic Properties of Acetylcysteine.The American Review of Respiratory Disease 1967; 96:962-970.
77. Hirsch SR and Kory RC.An Evaluation of the Effect of Nebulized N-Acetylcysteine on Sputum Consistency.J Allergy 1967; 39:265-273.
78. Dippy JE and Davis SS.Rheological Assessment of Mucolytic Agents on Sputum of Chronic Bronchitics.Thorax 1969; 24:707-713.
79. Rancourt RC, et al.Thioredoxin Liquefies and Decreases the Viscoelasticity of Cystic Fibrosis Sputum.American Journal of Physiology Lung Cellular and Molecular Physiology 2004; 286:L931-L938.
80. Shah PL, et al.In vivo Effects of Recombinant Human DNase I on Sputum in Patients with Cystic Fibrosis.Thorax 1996; 51:119-125.
81. Shak S, et al.Recombinant Human DNase I Reduces the Viscosity of Cystic Fibrosis Sputum.Proceedings of the National Academy of Sciences 1990; 87:9188-9192.
82. Bruce RA and quinton KC. Effect of Oral Alpha-Chymotrypsin on Sputum Viscosity.British Medical Journal 1962; 3:282-284.
83. Rawlins GA. Use of a Pancreatin-trypsin Solution for the Liquefaction of Sputa for Routine Bacteriological Examination.Journal of Clinical Pathology 1968; 21:531-532.
84. Gebre N, et al. Improved Microscopical Diagnosis of Pulmonary Tuberculosis in Developing Countries.Transactions of the Royal Society of Tropical Medicine and Hygiene 1995; 89:191-193.
85. Kaore NM, et al. Increased Sensitivity of Sputum Microscopy with Sodium Hypochlorite Concentration Technique: A Practical Experience at RNTCPCenter.Lung India; 28:17-20.
86. Anderson K, et al.Use and Abuse of Jungmann’s Method for the Liquefaction of Tuberculous Sputa.Journal of Clinical Pathology 1953; 6:124-127.
87. Farnia P, et al.Improving Sensitivity of Direct Microscopy for Detection of Acid-Fast Bacilli in Sputum: Use of Chitin in Mucus Digestion.J ClinMicrobiol2002; 40:508-511.
88. Subramanyam B, et al.An Alternative Sputum Processing Method Using Chitin for the Isolation of Mycobacterium tuberculosis.World Journal of Microbiology and Biotechnology 2010; 26:523-526.
89. Phillips BJ and Kaplan W.Effect of Cetylpyridinium Chloride on Pathogenic Fungi and Nocardiaasteroides in Sputum.J ClinMicrobiol1976;3:272-276.
90. Bobadilla-Del-Valle M, et al. Comparison of Sodium Carbonate, Cetyl-Pyridinium Chloride, and Sodium Borate for Preservation of Sputa for Culture of Mycobacterium tuberculosis.J ClinMicrobiol2003; 41:4487-4488.
91. Pardini M, et al.Cetyl-Pyridinium Chloride is Useful for Isolation of Mycobacterium tuberculosis from Sputa Subjected to Long-Term Storage.J ClinMicrobiol2005; 43:442-444.
92. SmithwickRW, et al.Use of Cetylpyridinium Chloride and Sodium Chloride for the Decontamination of Sputum Specimens That are Transported to the Laboratory for the Isolation of Mycobacterium tuberculosis.J ClinMicrobiol1975; 1:411-413.
93. Lieberman J andKurnick NB.The Induction of Proteolysis in Purulent Sputum by Iodides.Journal of Clinical Investigation 1964; 43:1892-1904.
94. Helb D, et al. Rapid Detection of Mycobacterium tuberculosis and Rifampin Resistance by Use of On-Demand, Near-Patient Technology.J ClinMicrobiol2010; 48:229-237.
95. Lorian V and Lacasse ML.N-Acetyl-L-Cysteine Sputum Homogenization and Its Mechanism of Action on Isolation of Tubercle Bacilli.Diseases of the Chest 1967;51:275-277.
96. Brady MF, et al. The MODS Method for Diagnosis of Tuberculosis and Multidrug Resistant Tuberculosis.Journal of Visualized Experiments 2008; 18: e845-e849.
97. WHO (World Health Organisation).Annex 7 - Decontamination methods.In: F.Portaels, P. Johnson and W.M Meyers (eds.), Buruli ulcer - Diagnosis of Mycobacterium ulcerans disease 2001;pp: 68-70.
98. SaraswathyVV, et al. Comparative Analysis of Cell Morphology in Sputum Samples Homogenized with Dithiothreitol, N-acetyl-l cysteine, Cytorich® Red Preservative and in Cellblock Preparations to Enhance the Sensitivity of Sputum Cytology for the Diagnosis of Lung Cancer.Diagnostic Cytopathology 2015; 43:551-558.
99. Baxter CG, et al.Homogenisation of Cystic Fibrosis Sputum by Sonication — An Essential Step for AspergillusPCR. Journal of Microbiological Methods 2011; 85:75-81.
100. KamgaHLF, et al.An Evaluation Study of the Sputum Smear Concentration Technique for the Laboratory Diagnosis of Pulmonary Tuberculosis.African Journal of Clinical and Experimental Microbiology 2010; 12:22-25.
101. GiacomelliLRB, et al.Improved laboratory safety by decontamination of unstained sputum smears for acid-fast microscopy. J ClinMicrobiol2005;43:4245-4248.
102. Hadad DJ, et al.Evaluation of Processing Methods to Equitably Aliquot Sputa for Mycobacterial Testing.J ClinMicrobiol2012; 50:1440-1442.
103. Pye A, et al. Simple Method for Quantifying Viable Bacterial Numbers in Sputum.Journal of Clinical Pathology 1995; 48:719-724.
104. Rantakokko-JalawaK and Jalawa J. Optimal DNA Isolation Method for Detection of Bacteria in Clinical Specimens by Broad-Range PCR.J ClinMicrobiol2002; 40:4211-4217.
105. Nauwelaers D, et al. Development of a Real-Time Multiplex RSV Detection Assay for Difficult Respiratory Samples, Using Ultrasone Waves and MNAzyme Technology.Journal of Clinical Virology 2009; 46:238-243.
106. Ulrich MP, et al.Evaluation of the Cepheid GeneXpert® system for detecting Bacillus anthracis.Journal of Applied Microbiology 2006; 100:1011-1016.
107. Deneer HG andKnight I.Inhibition of the Polymerase Chain Reaction by Mucolytic Agents.Clinical Chemistry 1994; 40:171-172.
108. DavisJL, et al. SecA1 PCR on Sputum or Oral Wash for the Diagnosis of Pulmonary Tuberculosis.Clinical Infectious Diseases 2009; 48:725-732.
109. Savic B, et al.Evaluation of Polymerase Chain Reaction, Tuberculostearic Acid Analysis, and Direct Microscopy for the Detection of Mycobacterium tuberculosis in Sputum.The Journal of Infectious Diseases 1992; 166:1177-1180.
110. Noordhoek GT, et al.Sensitivity and Specificity of PCR for Detection of Mycobacterium tuberculosis: a Blind Comparison Study among Seven Laboratories.J ClinMicrobiol1994; 32:277-284.
111. Eing BR, et al.Comparison of Roche CobasAmplicor Mycobacterium tuberculosis Assay with In-house PCR and Culture for Detection of M. tuberculosis. J ClinMicrobiol1998;36:2023-2029.
112. IchiyamaS, et al. Diagnostic Value of the Strand Displacement Amplification Method Compared to Those of Roche AmplicorPCR and Culture for Detecting Mycobacteria in Sputum Samples.J ClinMicrobiol1997; 35:3082-3085.
113. DavisJL, et al. Nucleic Acid Amplification Tests for Diagnosis of Smear-Negative in a HIV-Prevalence Setting: A Prospective Cohort Study.PLoS ONE 2011; 6:e16321.
114. Xiang X, et al.Comparison of Different Methods of Total RNA Extraction for Viral Detection in Sputum.Journal of Virological Methods 2001; 9:129-135.
115. Desjardin LE, et al. Alkaline Decontamination of Sputum Specimens Adversely Affects Stability of Mycobacterial mRNA.J ClinMicrobiol1996; 34:2435-2439.
116. Wilson S, et al.Concentration of Mycobacterium tuberculosis from Sputum Using Ligand-Coated Magnetic Beads.The International Journal of Tuberculosis and Lung Disease 2010; 14:1164-1168.
117. Liu J, et al.Increased Case Finding of Tuberculosis from Sputum and Sputum Deposits after Magnetic Bead Concentration of Mycobacteria.Journal of Microbiological Methods 2013; 93:144-147.
118. Wang X, et al.Bead Capture Increases the Sensitivity of Sputum Microscopy for the Diagnosis of Tuberculosis in Beijing, China.Transactions of the Royal Society of Tropical Medicine and Hygiene 2013; trt068.
119. Ghodbane R and Drancourt M.A Magnetic Bead Protocol for Culturing Mycobacterium tuberculosis from Sputum Specimens.J ClinMicrobiol 2013; 51:1578-1579.
120. Miörner H, et al.Improved Sensitivity of Direct Microscopy for Acid-Fast Bacilli: Sedimentation as an Alternative to Centrifugation for Concentration of Tubercle Bacilli.J ClinMicrobiol1996; 34:3206-3207.
121. PandeyA and asthana AK.Sedimentation Method, a Good Alternative to Centrifugation for Concentration of Acid Fast Bacilli in Developing Countries: A Preliminary Study from Western Uttar Pradesh. Indian J Med Microbiol2009; 27:83-84.