Pathogenesis

The pathogenesis of tuberculosis caused by organisms of the M. tuberculosis complex is discussed in Chapter 69. Inhalation of a single viable organism has been shown to lead to infection, although close contact is usually necessary. Of those who become infected with M. tuberculosis, 15% to 20% develop disease. The disease usually occurs some years after the initial infection, when the patient’s immune system breaks down for some reason other than the presence of tuberculosis bacilli in the lung. In a small percentage of infected hosts, the disease becomes systemic, affecting a variety of organs.

After ingestion of milk from infected cows, Mycobacterium bovis may penetrate the gastrointestinal mucosa or invade the lymphatic tissue of the oropharynx. An attenuated strain of M. bovis, bacillus Calmette-Guérin (BCG), has been used extensively in many parts of the world to immunize susceptible individuals against tuberculosis. Because mycobacteria are the classic examples of intracellular pathogens and the body’s response to BCG hinges on cell-mediated immunoreactivity, immunized individuals are expected to react more aggressively against all antigens that elicit cell-mediated immunity. In rare cases, an unfortunate individual’s immune system is so compromised that it cannot handle the BCG, and systemic BCG infection may develop.

Photochromogens

The photochromogens (Table 43-3) are slow-growing NTM that produce colonies that require light to form pigment.

Scotochromogens

The scotochromogens (Table 43-4) are slow-growing NTM that produce pigmented colonies whether grown in the dark or the light. The epidemiology of the potentially pathogenic scotochromogens has not been definitively described. In contrast to potentially pathogenic nonphotochromogens, these agents are rarely recovered in the clinical laboratory.

Nonphotochromogens

The nonphotochromogens (Table 43-5) are slow-growing NTM that produce unpigmented colonies whether grown in the dark or the light. Of the organisms in this group, M. terrae complex (M. terrae, M. triviale, and M. nonchromogenicum) and M. gastri are considered nonpathogenic for humans. The other nonphotochromogens are considered potentially pathogenic, and many are frequently recovered in the clinical laboratory. The nonphotochromogens belonging to Mycobacterium avium complex are frequently isolated in the clinical laboratory and are able to cause infection in the human host.

General Characteristics

The large group of organisms that constitute the RGM is divided into six major groups of potentially pathogenic species, based on pigmentation and molecular studies (see Box 43-1). Unlike the majority of other mycobacteria, most rapid-growers can grow on routine bacteriologic media and on media specific for cultivation of mycobacteria. On Gram staining, these organisms appear as weakly gram-positive rods resembling diphtheroids.

Epidemiology and Pathogenesis

The rapidly growing mycobacteria considered potentially pathogenic can cause disease in either healthy or immunocompromised patients. Like many other NTM, these organisms are ubiquitous in the environment and are present worldwide. They have been found in soil, marshes, rivers, and municipal water supplies (tap water) and in marine and terrestrial life forms. Infections caused by rapidly growing mycobacteria can be acquired in the community from environmental sources. They also can be nosocomial infections, resulting from medical interventions (including bone marrow transplantation), wound infections, and catheter sepsis. These organisms may be commensals on the skin. They gain entry into the host by inoculation into the skin and subcutaneous tissues as a result of trauma, injections, or surgery, or through animal contact.

The RGM also can cause disseminated cutaneous infections. The description of chronic pulmonary infections caused by rapidly growing mycobacteria suggests a possible respiratory route for acquisition of organisms present in the environment. Of the potentially pathogenic, rapidly growing NTM, M. fortuitum, M. chelonae, and M. abscessus are commonly encountered; these three species account for approximately 90% of clinical disease. Little is known about the pathogenesis of these organisms.

Spectrum of Disease

The spectrum of disease caused by the most commonly encountered rapid-growers is summarized in Table 43-6. The most common infection associated with RGM is posttraumatic wound infection. An increase in wound infections has been associated with planktonic M. abscessus, which can be identified as a rough colonial phenotype on artificial media; these organisms are capable of infecting macrophages. The smooth colonial phenotype typically is identified in biofilms and lacks infectivity.

General Characteristics

M. leprae has not yet been cultivated in vitro, although it can be cultivated in the armadillo and in the footpads of mice. Molecular biologic techniques have provided most of the information about this organism’s genomic structure and its various genes and their products. Although polymerase chain reaction (PCR) assays have been used to detect and identify M. leprae in infected tissues, the technique thus far has not proved as effective diagnostically as anticipated in indeterminate or paucibacillary (few organisms present) disease. Therefore, diagnosis of leprosy is based on distinct clinical manifestations, such as hypopigmented skin lesions and peripheral nerve involvement, in conjunction with a skin smear that tests positive for acid-fast bacilli.

Epidemiology and Pathogenesis

Understanding of the epidemiology and pathogenesis of leprosy is hampered by the inability to grow the organism in culture. In tropical countries, where the disease is most prevalent, it may be acquired from infected humans; however, infectivity is very low. Prolonged close contact and the host’s immunologic status play roles in infectivity.

Spectrum of Disease

Based on the host’s response, the spectrum of disease caused by M. leprae ranges from subclinical infection to intermediate stages of disease to full-blown and serious clinical manifestations involving the skin, upper respiratory system, testes, and peripheral nerves. The two major forms of the disease are a localized form, called tuberculoid leprosy, and a more disseminated form, called lepromatous leprosy. Patients with lepromatous leprosy are anergic to M. leprae because of a defect in their cell-mediated immunity. Because the organisms’ growth is unimpeded, these individuals develop extensive skin lesions containing numerous acid-fast bacilli; the organisms can spill over into the blood and disseminate. In contrast, individuals with tuberculoid leprosy do not have an immune defect, so the disease is localized to the skin and nerves; few organisms are observed in skin lesions. Most of the serious sequelae associated with leprosy are the result of this organism’s tropism for peripheral nerves.

Pulmonary Specimens

Pulmonary secretions may be obtained by any of the following methods: spontaneously produced or induced sputum, gastric lavage, transtracheal aspiration, bronchoscopy, and laryngeal swabbing. Most specimens submitted for examination are sputum, aerosol-induced sputum, bronchoscopic aspirations, or gastric lavage samples. Spontaneously produced sputum is the specimen of choice. To raise sputum, patients must be instructed to take a deep breath, hold it momentarily, and then cough deeply and vigorously. Patients must also be instructed to cover the mouth carefully while coughing and to discard tissues in an appropriate receptacle. Saliva and nasal secretions should not be collected, nor should the patient use oral antiseptics during the collection period. Sputum specimens must be free of food particles, residues, and other extraneous matter.

The aerosol (saline) induction procedure can best be done on ambulatory patients who are able to follow instructions. Aerosol-induced sputum specimens have been collected from children as young as 5 years of age. This procedure should be performed in an enclosed area with appropriate airflow. Operators should wear particulate respirators and take appropriate safety measures to prevent exposure. The patient is told that the procedure is being performed to induce coughing to raise sputum that the patient cannot raise spontaneously and that the salt solution is irritating. The patient is instructed to inhale slowly and deeply through the mouth and to cough at will, vigorously and deeply, coughing and expectorating into a collection tube. The procedure is discontinued if the patient fails to raise sputum after 10 minutes or feels any discomfort. Ten milliliters of sputum should be collected; if the patient continues to raise sputum, a second specimen should be collected and submitted. Specimens should be delivered promptly to the laboratory and refrigerated if processing is delayed.

Sputum collection guidelines recommend collection of an early morning specimen for 3 consecutive days. In many cases the third specimen demonstrates minimal recovery of organisms, and this collection may not be recommended in some laboratories. Pooled specimens are unacceptable because of an increased risk of contamination.

Gastric Lavage Specimens

Gastric lavage is used to collect sputum from patients who may have swallowed sputum during the night. The procedure is limited to senile, nonambulatory patients; children younger than 3 years of age (specimen of choice); and patients who fail to produce sputum by aerosol induction. The most desirable gastric lavage is collected at the patient’s bedside before the patient arises and before exertion empties the stomach. Gastric lavage cannot be performed as an office or clinic procedure.

The collector should wear a cap, gown, and particulate respirator mask and should stand beside (not in front of) the patient, who should sit up on the edge of the bed or in a chair, if possible. The Levine collection tube is inserted through a nostril, and the patient is instructed to swallow the tube. When the tube has been fully inserted, a syringe is attached to the end of the tube and filtered distilled water is injected into the tube. The syringe is then used to withdraw 5 to 10 mL of gastric secretions, which is expelled slowly down the sides of the 50-mL conical collecting tube. Samples should be adjusted to a neutral pH. The laboratory may choose to provide sterile receptacles containing 100 mg of sodium carbonate to reduce the acidity; this improves the recovery of organisms. The top of the collection tube is screwed on tightly, and the tube is held upright during prompt delivery to the laboratory. Three specimens should be collected over a period of consecutive days. Specimens should be processed within 4 hours.

Bronchial lavages, washings, and brushings are collected and submitted by medical personnel. These are the specimens of choice for detecting nontuberculous mycobacteria and other opportunistic pathogens in patients with immune dysfunction.

Urine Specimens

The incidence of urogenital infections shows little evidence of decreasing. About 2% to 3% of patients with pulmonary tuberculosis show urinary tract involvement, but 30% to 40% of patients with genitourinary disease have tuberculosis at some other site. The clinical manifestations of urinary tuberculosis, which are variable, include frequency of urination (most common), dysuria, hematuria, and flank pain. Definitive diagnosis requires recovery of acid-fast bacilli from the urine.

Early morning voided urine specimens (40 mL minimum) in sterile containers should be submitted daily for at least 3 days. The collection procedure is the same as for collecting a clean-catch midstream urine specimen (see Chapter 73). The 24-hour urine specimen is undesirable because of excessive dilution, higher contamination, and difficulty in concentrating. Catheterization should be used only if a midstream voided specimen cannot be collected.

Fecal Specimens

Acid-fast staining or culture of stool (or both) from patients with AIDS has been used to identify patients who may be at risk for developing disseminated M. avium complex disease. The clinical utility of this practice remains controversial; however, if screening stains and/or cultures are positive, dissemination often follows. Feces should be submitted in a clean, dry, wax-free container without preservative or diluent. Contamination with urine should be avoided.

Tissue and Body Fluid Specimens

Tuberculous meningitis is uncommon but occurs in both immunocompetent and immunosuppressed patients. A sufficient quantity of specimen is crucial for isolation of acid-fast bacilli from CSF. Very few organisms may be present in the spinal fluid, which makes their detection difficult. At least 10 mL of CSF is recommended for recovery of mycobacteria. Similarly, as much as possible of other body fluids (10 to 15 mL minimum), such as pleural, peritoneal, and pericardial fluids, should be collected in a sterile container or syringe with a Luer-tip cap. Tissues may be immersed in saline or wrapped in gauze. Swabs are discouraged, because the recovery of organisms is decreased.

Blood Specimens

Immunocompromised patients, particularly those infected with HIV, can have disseminated mycobacterial infection; most of these infections are caused by M. avium complex. A blood culture positive for MAC is always associated with clinical evidence of disease. Recovery of mycobacteria is improved with blood collection in either a broth or the Isolator lysis-centrifugation system (see Chapter 68). Some studies have indicated that the lysis-centrifugation system is advantageous, because quantitative data can be obtained with each blood culture; in patients with AIDS, quantitation of such organisms can be used to monitor therapy and determine the prognosis. However, the necessity of quantitative blood cultures remains unclear.

Blood for culture of mycobacteria should be collected as for routine blood cultures. Blood collected in regular phlebotomy procedures in anticoagulants such as sodium polyanethol sulfonate (SPS), heparin, and citrate may be used to inoculate cultures for the recovery of Mycobacterium species. Conventional blood culture collection systems are unacceptable for the isolation of Mycobacterium spp. However, specialized automated systems are available for growth of Mycobacterium spp., including the Bactec MGIT 960 system (Becton-Dickinson, Franklin Lakes, N.J.), and the BacT/ALERT 3D (Biomerieux, Durham, N.C.).

Wounds, Skin Lesions, and Aspirates

An aspirate is the best type of specimen for culturing of a skin lesion or wound. The skin should be cleansed with alcohol before aspiration of the material into a syringe. If the volume is insufficient for aspiration, pus and exudates may be obtained on a swab and then placed in a transport medium, such as Amie’s or Stuart’s medium (dry swabs are unacceptable). However, a negative culture of a specimen obtained on a swab is not considered reliable, and this should be noted in the culture report.

Contaminated Specimens

Most specimens submitted for mycobacterial culture consist of organic debris, such as mucin, tissue, serum, and other proteinaceous material contaminated with organisms. A typical example of such a specimen is sputum. Laboratories must process these specimens to kill or reduce contaminating bacteria that can rapidly outgrow mycobacteria, and mycobacteria are released from mucin and/or cells. After decontamination, mycobacteria are concentrated, usually by centrifugation, to enhance their detection by acid-fast stain and culture. Unfortunately, there is no single ideal method for decontaminating and digesting clinical specimens. Although continuously faced with the inherent limitations of various methods, laboratories must strive to maximize the survival and detection of mycobacteria while maximizing the elimination of contaminating organisms. Rapidly growing mycobacteria are especially susceptible to high or prolonged exposure to greater than or equal to 2% sodium hydroxide (NaOH). Digestion-decontamination procedures should be as gentle as possible.

Inadequate Specimens and Rejection Criteria

Identification and detection of Mycobacterium spp. is costly and time consuming. It is essential that the laboratory have a detailed policy regarding the rejection of inadequate specimens for the identification of these organisms. Specimens should be rejected according to the following guidelines: (1) insufficient volume, (2) contamination with saliva, (3) dried swabs, (4) pooled sputum or urine, (5) container has been compromised, broken or leaking, and (6) length of time from collection to processing is too long.

Specimens Not Requiring Decontamination

Tissues or body fluids collected aseptically usually do not require the digestion and decontamination methods used with contaminated specimens. The processing of clinical specimens that do not routinely require decontamination for acid-fast culture is described here. If such a specimen appears contaminated because of color, cloudiness, or foul odor, Gram staining is performed to detect bacteria other than acid-fast bacilli. Specimens found to be contaminated should be processed as described in the preceding section.

CSF should be handled aseptically and centrifuged for 30 minutes at 3600× g to concentrate the bacteria. The supernatant is decanted, and the sediment is vortexed thoroughly before the smear is prepared and the media inoculated. If insufficient quantity of spinal fluid is received, the specimen should be used directly for smear and culture. Recovery of acid-fast bacilli from CSF is difficult, and additional solid or liquid media should be inoculated if material is available.

Pleural fluid should be collected in sterile anticoagulant (1 mg/mL ethylenediaminetetraacetic acid [EDTA] or 0.1 mg/mL heparin). If the fluid becomes clotted, it should be liquefied with an equal volume of sputolysin and vigorously mixed. To lower the specific gravity and density of pleural fluid, 20 mL is transferred to a sterile, 50-mL centrifuge tube, and the specimen is diluted by filling the tube with distilled water. The tube is inverted several times to mix the suspension and then centrifuged at 3600× g for 30 minutes. The supernatant should be removed, and the sediment should be suspended for smear and culture.

Joint fluid and other sterile exudates can be handled aseptically and inoculated directly to media. Bone marrow aspirates may be injected into Pediatric Isolator tubes (Alere, Waltham, MA), which help prevent clotting; the specimen can be removed with a needle and syringe for preparation of smears and cultures. As an alternative, these specimens are either inoculated directly to media or, if clotted, treated with sputolysin or glass beads and distilled water before concentration.

Acid-Fast Stains

The cell walls of mycobacteria contain long-chain, multiply cross-linked fatty acids, called mycolic acids. Mycolic acids probably complex basic dyes, contributing to the characteristic of acid-fastness that distinguishes mycobacteria from other bacteria. Mycobacteria are not the only group with this unique feature. Species of Nocardia and Rhodococcus are also partially acid-fast; Legionella micdadei, a causative agent in pneumonia, is partially acid-fast in tissue. Cysts of the genera Cryptosporidium and Isospora are distinctly acid-fast. The mycolic acids and lipids in the mycobacterial cell wall probably account for the unusual resistance of these organisms to the effects of drying and harsh decontaminating agents in addition to the property of acid-fastness.

When Gram stained, mycobacteria usually appear as slender, poorly stained, beaded, gram-positive bacilli (Figure 43-2); sometimes they appear as “gram neutral,” or “gram-ghosts,” by failing to take up either crystal violet or safranin. Acid-fastness is affected by the age of colonies, the medium on which growth occurs, and exposure to ultraviolet light. Rapidly growing species appear to be acid-fast variable.

Three types of staining procedures are used in the laboratory for rapid detection and confirmation of acid-fast bacilli: fluorochrome, Ziehl-Neelsen, and Kinyoun. Smears for all methods are prepared in the same way (see Procedure 43-2 on the Evolve site).

Visualization of acid-fast bacilli in sputum or other clinical material should be considered only presumptive evidence of tuberculosis, because staining does not specifically identify M. tuberculosis. The report form should indicate this. For example, M. gordonae, a nonpathogenic scotochromogen commonly found in tap water, has been a problem when tap water or deionized water has been used in the preparation of smears or even when patients have rinsed their mouths with tap water before using an aerosolized saline solution to induce sputum. However, the incidence of false-positive smears is very low when good quality control is maintained. Conversely, acid-fast–stained smears of clinical specimens require at least 10⁴ acid-fast bacilli per milliliter for detection from concentrated specimens.

Methods

Fluorochrome Stain.

Fluorochrome staining is the screening procedure recommended for laboratories that have a fluorescent (ultraviolet) microscope (see Procedure 43-3 on the Evolve site). Fluorochrome stain is more sensitive than the conventional carbolfuchsin stains, because the fluorescent bacilli stand out brightly against the background (Figure 43-3). Because the smear can be examined initially at lower magnifications (×250 to ×400), more fields can be visualized in a short period. In addition, a positive fluorescent smear may be restained using the conventional Ziehl-Neelsen or Kinyoun procedure, thereby saving the time needed to make a fresh smear. Screening of specimens with rhodamine or rhodamine-auramine results in a higher yield of positive smears and substantially reduces the time needed to examine smears.

One drawback of the fluorochrome stains is that many rapid-growers may not appear fluorescent with these reagents. All positive fluorescent smears should be confirmed with a Ziehl-Neelsen stain or by examination by another technologist. It is important to wipe the immersion oil from the objective lens after examining a positive smear, because stained bacilli can float off the slide into the oil, possibly contributing to a false-positive reading for the next smear examined.

Procedure 43-3   Auramine-Rhodamine Fluorochrome Stain

Principle

The fluorochrome dyes used in this stain complex to the mycolic acids in acid-fast cell walls. Detection of fluorescing cells is enhanced by the brightness against a dark background.

Method

1. Heat-fix slides at 80° C for at least 15 minutes or for 2 hours at 65° to 70° C.

2. Flood slides with auramine-rhodamine reagent and allow to stain for 15 to 20 minutes at room temperature.

3. Rinse with deionized water and tilt slide to drain.

4. Decolorize with 0.5% acid-alcohol (70% ethanol and 0.5% hydrochloric acid) for 2 to 3 minutes.

5. Rinse with deionized water and tilt slide to drain.

6. Flood slides with 0.5% potassium permanganate for 2 to 4 minutes.

7. Rinse with deionized water and air dry.

8. Examine under low power (250×) for fluorescence.

Expected Results

Mycobacterium spp. fluoresce yellow to orange, depending on the filter system used.

Fuchsin Acid-Fast Stains.

The classic carbolfuchsin stain (Ziehl-Neelsen) requires heating of the slide for better penetration of the stain into the mycobacterial cell wall; hence, it is also known as the hot stain procedure (see Procedure 6-3 on the Evolve site). With Ziehl-Neelsen staining, Mycobacterium spp. appear red or have a red-blue, beaded appearance, whereas nonmycobacteria appear blue.

Procedure 6-4, which can be found on the Evolve site, describes the Kinyoun acid-fast stain. The method is similar to Ziehl-Neelsen staining, but no heat is used (see Figure 43-3); this technique is known as the cold stain procedure. If present, typical acid-fast bacilli appear as purple to red, slightly curved, short or long rods (2 to 8 µm); they also may appear beaded or banded (M. kansasii). For some nontuberculous species, such as M. avium complex, they appear pleomorphic, usually coccoid.

Antigen-Protein Detection

The detection of microbial products or components has been used in recent years to diagnose infections caused by M. tuberculosis. For example, tuberculostearic acid is a fatty acid that can be extracted from the cell wall of mycobacteria and detected by gas chromatography or mass spectrometry in clinical samples containing few mycobacteria. Because of the limited number of species that can cause meningitis and because M. tuberculosis appears to be the only one of these species that releases tuberculostearic acid into the surrounding environment, the presence of this substance in CSF is thought to be diagnostic of tuberculous meningitis. Performance of this assay is limited to a few laboratories. Various immunoassays for antigen detection directly in clinical specimens, including sputum and CSF, have been evaluated and show some promise.

Production of adenosine deaminase, a host enzyme, is increased in certain infections caused by M. tuberculosis. For example, elevated levels of this enzyme were found in most patients with tuberculous pleural effusions (98% sensitive); the test for the enzyme also was determined to be highly specific (96% specificity).

Immunodiagnostic Testing

As previously discussed, interferon-gamma release assays have become more widely used for the diagnosis of tuberculosis. The available test systems, T-SPOT-TB (Oxford Immunotec, Oxford, United Kingdom) and QuantiFERON Gold In-Tube (QFNG-IT; Cellestis, Chadstone, Victoria, Australia), do not typically cross react with nontuberculous mycobacterium, are not affected by the BCG vaccine, and are not as variable as the historical serologic tuberculin skin tests. The T-SPOT-TB assay is an enzyme-linked immunospot assay that requires isolation and incubation of peripheral blood mononuclear cells (PBMCs). It takes approximately 2 days and is technically complicated. The QFNG-IT assay measures the stimulation of T-cell interferon-gamma in whole blood in a tube precoated with M. tuberculosis antigens. It yields results in approximately 8 hours. Neither assay distinguishes between latent and active infections. In addition, specificity and sensitivity vary in the population tested, including immunocompromised patients and children. Variation is associated with the patient’s CD4 cell count; therefore, interpretations and results should be evaluated with caution.

Genetic Sequencing and Nucleic Acid Amplification

Subsequent to the introduction of commercially available hybridization assays, commercially available and in-house–developed nucleic acid amplification tests were used successfully for early identification of M. tuberculosis complex grown in liquid cultures. Currently, PCR-based sequencing for mycobacterial identification consists of PCR amplification of mycobacterial DNA with genus-specific primers and sequencing of the amplicons. The organism is identified by comparison of the nucleotide sequence with reference sequences. The most reliable sequence for identification of mycobacteria is the approximately 1500 bp 16S rRNA gene. However, only a 600 bp sequence at the 5′ end is required for identification. The sequence homogeneity in the M. tuberculosis complex prevents the use of this sequence to differentiate these species. This region contains both conserved and variable regions, which makes it an ideal target for identification purposes.

Despite the accuracy of PCR-based sequencing to identify mycobacteria, problems remain: the sequences in some databases are not accurate; no present consensus exists as to the quantitative definition of a genus or species based on 16S rRNA gene sequence data; and procedures are not standardized. In addition, the 16S rRNA 5′ region contains two hypervariable regions, A and B. The A region provides the signature sequences for species identification. However, M. chelonae and M. abscessus both require additional sequencing, because the A and B regions are identical and the 3′ end of the 16srRNA contains a 4-bp sequence difference.

Several other genes have also been used to identify mycobacterial species, including the 23S rRNA, ITS 1, hsp65, rpoB, and gyrB gene. The 23S rRNA sequence is 3100 bp in length, which limits accurate sequencing. ITS 1 is a spacer sequence located between the 16S and 23S rRNA genes. This sequence, which is only 200 to 330 bp, is more easily analyzed. The limitation of this sequence is that it is not a genus-specific sequence; therefore, results may be affected by contaminating bacteria. The 65 kDa heat shock protein, also referred to as the groEL2 gene, is a 440-bp fragment that can be amplified and analyzed with restriction digestion, followed by agarose electrophoresis. The hsp65 is highly conserved but contains a greater variation in polymorphisms than the 16S RNA, particularly in a 441-bp region referred to as the “Telanti fragment.” This allows for differentiation of Mycobacterium species based on the variation in restriction fragment length polymorphisms (RFLPs). Repetitive sequence–based PCR, Diversilab (Biomérieux, Durham, N.C.), demonstrates better species discrimination than RFLP. In addition, a commercially available system in which the 16S to 23S rRNA spacer region of mycobacterial species (INNO-LiPA Mycobacteria; Innogenetics, Ghent, Belgium) has been successfully used to directly detect and identify several of the most clinically relevant mycobacterial species in aliquots of positive liquid culture. However, caution should be used in interpretation of results, because some cross reactivity has occurred with closely related species.

Another commercial system, GenoType Mycobacterium (Hain Lifescience GmbH, Nehrin, Germany), which uses a similar format, has additional probes from M. celatum, M. malmoense, M. peregrinum, M. phlei, and two subgroups of M. fortuitum, in addition to a supplemental kit that allows for 16 additional mycobacterial species. Yet another commercial system, MicroSeq500 16S rRNA (Applied Biosystems, Foster City, California), sequences a 500-bp region and uses a comparative database for species identification.

The rpoB gene encodes the beta-subunit in the organism’s RNA polymerase. Mutations in this gene confer rifampin resistance to M. tuberculosis. Different regions in this gene have been used to identify rapid-growing isolates, but little data are available for the slow-growing species. Finally, the gyrB gene encodes the beta-subunit in the organism’s topoisomerase II. Several single nucleotide polymorphisms have been identified in this gene that are useful in distinguishing species in the M. tuberculosis complex. After amplification, identification and differentiation of species requires restriction analysis and gel electrophoresis.

Additional molecular techniques, such as conventional and real-time PCR, have been used to detect M. tuberculosis directly in clinical specimens. For example, the Amplicor Mycobacterium tuberculosis test (Roche Diagnostic Systems, Branchburg, New Jersey) uses PCR to detect M. tuberculosis directly in respiratory specimens. The Amplified Mycobacterium tuberculosis Direct Test (AMTD; Gen-Probe, San Diego, California) is based on ribosomal RNA amplification. The Roche assay currently is approved by the FDA for use only on acid-fast, smear-positive specimens, because numerous studies have demonstrated less than optimum sensitivity on smear-negative specimens. Because of subsequent kit modifications that improved sensitivity, Gen-Probe’s assay is approved on both smear-positive and smear-negative specimens. Ribosomal RNA is released from the mycobacteria by means of a lysing agent, sonication, and heat. The specific DNA probe is allowed to react with the extracted rRNA to form a stable DNA-RNA hybrid. Any nonhybridized DNA–acridinium ester probes are chemically degraded. When an alkaline hydrogen peroxide solution is added to elicit chemiluminescence, only the hybrid-bound acridinium ester is available to emit light; the amount of light emitted is directly related to the amount of hybridized probe. The light produced is measured on a chemiluminometer. Numerous laboratories have incorporated these tests into their routine procedures.

The Amplicor test (Roche Diagnostic Systems), which uses TaqMan technology and is a real-time PCR test, has received FDA approval for smear-positive respiratory specimens from patients suspected of having tuberculosis. These tests are limited in the number of species they are able to identify. Clinical laboratories have developed their own PCR assays to detect M. tuberculosis directly in clinical specimens.

Line probe assays (DNA strip assays) involve PCR amplification coupled with a reverse hybridization step. The target sequence is amplified using biotinylated primers. The amplicon is then hybridized to membrane-immobilized, sequence-specific probes for each species. The membrane is developed using an enzyme-mediated reaction and color indicator to analyze the banding pattern. Banding patterns are species specific based on the immobilized probe map on the membrane. A commercially available line probe assay (GenoType MTBC; Hain Lifescience, Nehren, Germany) enables the identification of M. tuberculosis complex organisms at the species level using the 23S rRNA.

In addition to assays developed in-house and the Genotype MTBC, five non-FDA-approved commercial amplification tests are widely used outside the United States. The Artus M. tuberculosis PCR kit (Qiagen GmbH, Hilden/Hamburg, Germany) assay uses real-time PCR for amplification of the 16S rRNA gene; the ProbeTec Direct TB energy transfer system (Becton Dickinson, Sparks, Maryland) uses strand displacement amplification technology; the RealArt M. tuberculosis TM PCR reagents (Abbott Laboratories, Abbott Park, Illinois) is a real-time PCR assay using the ABI Prism 7000 system; and the Loop-mediated isothermal amplification test (Eiken Chemical, Tokyo) uses isothermal amplification and UV light detection. These systems have sensitivities and specificities comparable to those of the FDA-approved amplification assays. The GeneXpert system (Cepheid, Sunnyvale, California), which is used for real-time PCR detection of M. tuberculosis complex and resistance to rifampin, uses amplification of the rpoB gene previously discussed in this section.

Currently no molecular assays are available for direct detection of nontuberculous mycobacteria. In 2004, the Centers for Disease Control established a national tuberculosis genotyping system. Details and updates are available at http://www.cdc.gov/tb/programs/default.htm.

Chromatographic Analysis

Analysis of mycobacterial lipids by chromatographic methods, including thin-layer chromatography, gas-liquid chromatography (GLC), capillary gas chromographic methods, and reverse-phase high-performance liquid chromatography (HPLC), has been used to identify mycobacteria. In HPLC, a liquid mobile phase is combined with various technical advances to separate large cellular metabolites and components. HPLC of extracted mycobacteria is a specific and rapid method for identifying species. Many state health departments and the CDC now use this method routinely. The long-chain mycolic acids are separated better by HPLC than by GLC, because they do not withstand the high temperatures needed for GLC. The patterns produced by different species are very easily reproducible, and a typical identification requires only a few hours.

Cultivation

A combination of different culture media is required to optimize recovery of mycobacteria from culture; at least one solid medium in addition to a liquid medium should be used. The ideal media combination should be economical and should support the most rapid and abundant growth of mycobacteria, allow for the study of colony morphology and pigment production, inhibit the growth of contaminants.

Solid Media

Solid media, such as those listed in Box 43-3, are recommended because of the development of characteristic, reproducible colonial morphology, good growth from small inocula, and a low rate of contamination. Optimally, at least two solid media (a serum [albumin] agar base medium, [e.g., Middlebrook 7H10] and an egg-potato base medium [e.g., Löwenstein-Jensen, or L-J]) should be used for each specimen (these media are available from commercial sources). All specimens must be processed appropriately before inoculation. It is imperative to inoculate test organisms to commercially available products for quality control (see Procedure 43-4 on the Evolve site).

Cultures are incubated at 35° C in the dark in an atmosphere of 5% to 10% carbon dioxide (CO₂) and high humidity. Tube media are incubated in a slanted position with screw caps loose for at least 1 week to allow for evaporation of excess fluid and the entry of CO₂; plated media are either placed in a CO₂-permeable plastic bag or wrapped with CO₂-permeable tape. If specimens obtained from the skin or superficial lesions are suspected to contain M. marinum or M. ulcerans, an additional set of solid media should be inoculated and incubated at 25° to 30° C. In addition, a chocolate agar plate (or placement of an X-factor [hemin] disk on conventional media) and incubation at 25° to 33° C is needed for recovery of M. haemophilum from these specimens. RGM optimally require incubation at 28° to 30° C.

Cultures are examined weekly for growth. Contaminated cultures are discarded and reported as “contaminated, unable to detect presence of mycobacteria”; additional specimens are also requested. If available, sediment may be recultured after enhanced decontamination or by inoculating the sediment to a more selective medium. Most isolates appear between 3 and 6 weeks; a few isolates appear after 7 or 8 weeks of incubation. When growth appears, the rate of growth, pigmentation, and colonial morphology are recorded. The typical colonial appearance of M. tuberculosis and other mycobacteria is shown in Figure 43-4. After 8 weeks of incubation, negative cultures (those showing no growth) are reported, and the cultures are discarded.

Because of the resurgence of tuberculosis in the United States in the late 1980s and early 1990s, significant effort has been put into developing methods to provide more rapid diagnosis of tuberculosis. Welch et al.² refined a method that reduced the time to detection of mycobacterial growth by half or more, compared with conventional culture methods, by using a thinly poured Middlebrook 7H11 plate. These plates are inoculated in a routine manner, sealed, incubated, and examined microscopically (×40) at regular intervals for the appearance of microcolonies. Presumptive identification of M. tuberculosis or M. avium complex could be made for about 83% of the isolates within 10 and 11 days after inoculation, respectively.

Liquid Media

In general, use of a liquid media system reduces the turnaround time for isolation of acid-fast bacilli to approximately 10 days, compared with 17 days or longer for conventional solid media. Several different systems are available for culturing and detecting the growth of mycobacteria in liquid media. The most commonly used systems are summarized in Table 43-8. Growth of mycobacteria in liquid media, regardless of the type, requires 5% to 10% CO₂; CO₂ is either already provided in the culture vials or is added according to the manufacturer’s instructions. When growth is detected in a liquid medium, acid-fast staining of a culture aliquot is performed to confirm the presence of acid-fast bacilli, and the material is subcultured to solid agar. Gram staining can also be performed if contamination is suspected.

Interpretation

Although isolation of MAC organisms indicates infection, the clinician must determine the clinical significance of isolating NTM in most cases; in other words, does the organism represent mere colonization or significant infection? Because these organisms vary greatly in their pathogenic potential, can colonize an individual without causing infection, and are ubiquitous in the environment, interpretation of a positive NTM culture is complicated. Therefore, the American Thoracic Society has recommended diagnostic criteria for NTM disease to help physicians interpret culture results.

Conventional Phenotypic Tests

Biochemical Testing.

Once categorized into a preliminary subgroup based on its growth characteristics, an organism must be definitively identified to species or complex level. Although conventional biochemical tests can be used for this purpose, new methods (discussed later in this section) have replaced biochemical tests for identifying mycobacterial species because of the previously discussed limitations of phenotypic testing. Although key biochemical tests are still discussed in this edition, the reader must be aware that this approach to identification ultimately will be replaced by molecular methods. Table 43-10 summarizes distinctive properties of the more commonly cultivable mycobacteria isolated from clinical specimens; key biochemical tests for each of the major mycobacterial groupings, including M. tuberculosis complex, are listed in Table 43-11. The following sections address key biochemical tests.

TABLE 43-10

Distinctive Properties of Commonly Cultivable Mycobacteria Encountered in Clinical Specimens

Group/Complex	Species	Optimal Temp (°C)	Usual Colonial Morphologya	Niacin	Growth on TCH (10 mg/mL)b	Nitrate Reduction	Semi quantitative Catalase (>45 mm)	68° C Catalase	Tween Hydrolysis, 5 Days	Tellurite Reduction	Tolerance to 5% NaCl	Arylsulfatase, 3 Days	Iron Uptake	Growth on MacConkey Agar	Urease	Pyrazinamidase, 4 Days
M. tuberculosis complex	M. tuberculosis	37	R	+	+	+	−	−	−c		−	−	−	−	±	+
	M. bovis	37	Rt	−	−	−	−	−	−		−	−	−	−	±	−
	M. africanum	37	R	V	V	V	−	−	−	−	−	−	−	−	+	−
Photochromogens	M. marinum	30	S/SR		+	−	−	−	+		−	−	d	−	−/+	+
	M. kansasii	35	SR/S	−	+	+	+	+	+		−	−	−	−	+	−
	M. simiae	37	S	±	+	−	+	+	−	+	−	−	−		±	+
	M. asiaticum	37	S	−	+	−	+	+	+	−	−	−	−		−	−
Scotochromogens	M. scrofulaceum	37	S	−	+	−	+	+	−		−	−	V	−	V	±
	M. szulgai	37	S or R	−	+	+	+	+	c	±	−	−	V	−	+	+
	M. gordonae	37	S	−	+	−	+	+	+	−	−	−	V	−	V	±
Nonphotochromogens	M. avium complex	35-37	St/R	−	+	−	+	±	−	+	−	−	−		−	+
	M. genavensee	37	St	−	+	−	−	+	+			−	−		+	+
	M. gastri	35	S/SR/R	−	+	−	+	−	+		−	−	−	−		−
	M. malmoense	30	S	−	+	−	−	±	+	+	−	−	−		−	+
	M. haemophilumf	30	R	−	+	−	−	−	−	−	−	−	−	−	−	+
	M. shimoidei	37	R	−	+	−	−	−	+		−	−	−	−	−	+
	M. ulcerans	30	R	−	+	−	−	+	−		−		−		−	−
	M. flavescensg	37	S	−	+	+	+	+	+		+	−	−	−	+	+
	M. xenopih	42	Sf	−	+	−	−	+	−		−	−	±	−	−	V
	M. terrae complex M. terrae M. trivialei M. nonchromogenicum	35	SR	−	−	+	+	+	+	−	−	−	−	V	−	V
Rapidly growing	M. fortuitum group	28-30	Sf/Rf	−	+	+	+	+	V	+	+	+	+	+	+	+
	M. chelonae	28-30	S/R	−/+	+	−	+	V	V	+	−	−	+	+	+	+
	M. abscessus	28-30	S/R	−		−	+	V	V	+	+	−	+	+	+
	M. smegmatis	28-30	R/S	−	+	+	+	+	+	+	+	+	−	−

Plus and minus signs indicate the presence or absence, respectively, of the feature; blank spaces indicate either that the information is not currently available or that the property is unimportant.

V, Variable; ±, usually present; , usually absent.

See Versalovic J: Manual of clinical microbiology, ed 10, Washington, DC, 2011, ASM Press, for biochemical reactions of other mycobacterial species and for additional biochemical reactions on the mycobacteria included in this table.

^aR, Rough; S, smooth; SR, intermediate in roughness; t, thin or transparent; f, filamentous extensions.

^bTCH, Thiophene-2-carboxylic acid hydrazide.

^cTween hydrolysis may be positive at 10 days.

^dArylsulfatase, 14 days, is positive.

^eRequires mycobactin for growth on solid media.

^fRequires hemin as a growth factor.

^gYoung cultures may be nonchromogenic or have only pale pigment that may intensify with age.

^hStrains of M. xenopi can be nonphotochromogenic or scotochromogenic.

ⁱM. triviale is tolerant to 5% NaCl, and a rare isolate may grow on MacConkey agar.

TABLE 43-11

Key Biochemical Reactions to Help Differentiate Organisms Belonging to the Same Mycobacterial Group

Mycobacterial Group	Key Biochemical Tests
M. tuberculosis complex	Niacin, nitrate reduction; susceptibility to thiophene-2-carboxylic acid hydrazide (TCH) if M. bovis is suspected
Photochromogens	Tween 80 hydrolysis, nitrate reduction, pyrazinamidase, 14-day arylsulfatase, urease, niacin
Scotochromogens	Permissive growth temperature, Tween 80 hydrolysis, nitrate reduction, semiquantitative catalase, urease, 14-day arylsulfatase
Nonphotochromogens	Heat-resistant and semiquantitative catalase activity, nitrate reduction, Tween 80 hydrolysis, urease, 14-day arylsulfatase, tellurite reduction, acid phosphatase activity
Rapidly growing	Growth on MacConkey agar, nitrate reduction, Tween 80 hydrolysis, 3-day arylsulfatase, iron uptake

Niacin.

Niacin (nicotinic acid) plays an important role in the oxidation-reduction reactions that occur during mycobacterial metabolism. Although all species produce nicotinic acid, M. tuberculosis accumulates the largest amount. (M. simiae and some strains of M. chelonae also produce niacin.) Niacin therefore accumulates in the medium in which these organisms are growing. A positive niacin test (see Procedure 43-6 on the Evolve site) is preliminary evidence that an organism that exhibits a buff-colored, slow-growing, rough colony may be M. tuberculosis (Figure 43-6). However, this test is not sufficient to confirm identification. If sufficient growth is present on an initial L-J slant (the egg-base medium enhances accumulation of free niacin), a niacin test can be performed immediately. If growth on the initial culture is scant, the subculture used for growth rate determination can be used. If this culture yields only rare colonies, the colonies should be spread around with a sterile cotton swab (after the growth rate has been determined) to distribute the inoculum over the entire slant. The slant then is incubated until light growth over the surface of the medium is visible. For reliable results, the niacin test should be performed only from cultures on L-J medium that are at least 3 weeks old and show at least 50 colonies; otherwise, enough detectable niacin might not have been produced.

Procedure 43-6   Niacin Test Performed with Commercially Available Filter Paper Test Strips*

Principle

The accumulation of niacin in an egg-based medium, the result of lack of an enzyme that converts niacin to another metabolite in the coenzyme pathway, is characteristic of M. tuberculosis and a few other species. Niacin is measured by a colored end product.

Method

1. Add 1 mL of sterile distilled water to the surface of the egg-based medium on which the colonies to be tested are growing.

2. Lay the tube horizontally so that the fluid is in contact with the entire surface. Using a pipette, scratch or lightly poke through the surface of the agar; this allows the niacin in the medium to dissolve in the water.

3. Allow the tube to sit for up to 30 minutes at room temperature. It can incubate longer to achieve a stronger reaction.

4. Remove 0.6 mL of the distilled water (which appears cloudy at this point) to a clean, 12 × 75 mm screw cap or snap top test tube. Insert a niacin test strip with the arrow down, following the manufacturer’s instructions.

5. Cap the tube tightly and incubate at room temperature, occasionally shaking the tube to mix the fluid with the reagent on the bottom of the strip.

6. After 20 minutes, observe the color of the liquid against a white background (see Figure 45-6).

Expected Results

Yellow liquid indicates a positive test result. The color of the strip should not be considered when evaluating results. If the liquid is clear, the test result is negative. Discard the strip into an alkaline disinfectant (10% sodium hydroxide [NaOH]) to neutralize the cyanogen bromide.

---

^*Manufactured by Remel, Lenexa, Kansas.

Nitrate Reduction.

This test is valuable for identifying M. tuberculosis, M. kansasii, M. szulgai, and M. fortuitum. The ability of acid-fast bacilli to reduce nitrate is influenced by the age of the colonies, temperature, pH, and enzyme inhibitors. Although rapid-growers can be tested within 2 weeks, slow-growers should be tested after 3 to 4 weeks of luxuriant growth. Commercially available nitrate strips yield acceptable results only with strongly nitrate-positive organisms, such as M. tuberculosis. This test may be tried first because of its ease of performance. The M. tuberculosis–positive control must be strongly positive in the strip test, or the test results are unreliable. If the paper strip test is negative or if the control test result is not strongly positive, the chemical procedure (see Procedure 43-7 on the Evolve site) must be carried out using strong and weakly positive controls.

Procedure 43-7   Nitrate Reduction Test Using Chemical Reagents

Principle

As in the conventional nitrate test, the presence of nitrite (the product of the nitroreductase enzyme) is detected by a red product that results with the addition of several reagents. If the enzyme has reduced nitrate past nitrite to gas, then addition of zinc dust (which converts nitrate to nitrite) will detect the lack of nitrate in the reaction medium.

Method

1. Prepare the dry crystalline reagent as follows:

Sulfanilic acid (Sigma Chemical Co., St Louis, Mo.)	1 part
N-(1-Naphthyl)ethylenediamine dihydrochloride (Eastman Chemical Co., Rochester, N.Y.)	1 part
L-tartaric acid (Sigma Chemical Co.)	10 parts

These crystals can be measured with any small scoop or tiny spoon, because the proportions are by volume, not weight. The mixture should be ground in a mortar and pestle to ensure adequate mixing, because the crystals have different textures. The reagent can be stored in a dark glass bottle at room temperature for at least 6 months.

2. Add 0.2 mL sterile distilled water to a 16 × 125 mm screw cap tube. Emulsify two very large clumps of growth from a 4-week culture on Löwenstein-Jensen agar in the water. The suspension should be milky.

3. Add 2 mL nitrate substrate broth (Becton-Dickinson, Franklin Lakes, N.J. or Remel, Lenexa, Kansas) to the suspension and cap tightly. Shake gently and incubate upright for 2 hours in a 35° C water bath.

4. Remove from water bath and add a small amount of the crystalline reagent. A wooden stick or a small spatula can be used to add crystals; the amount is not critical. Examine immediately.

Expected Results

Development of a pink to red color indicates the presence of nitrite, demonstrating the organism’s ability to reduce nitrate to nitrite. If no color results, the organisms may have reduced nitrate beyond nitrite (as in the conventional nitrate test). Add a small amount of powdered zinc to the negative tube. The development of a red color indicates that unreduced nitrate was present in the tube and the organism was nitroreductase negative.

Catalase.

Most species of mycobacteria, except for certain strains of M. tuberculosis complex (some isoniazid-resistant strains) and M. gastri, produce the intracellular enzyme catalase, which splits hydrogen peroxide into water and oxygen. Catalase can be assessed by using the semiquantitative catalase test or the heat-stable catalase test.

• The semiquantitative catalase test is based on the relative activity of the enzyme, as determined by the height of a column of bubbles of oxygen (Figure 43-7) formed by the action of untreated enzyme produced by the organism. Based on the semiquantitative catalase test, mycobacteria are divided into two groups: those that produce less than 45 mm of bubbles and those that produce more than 45 mm of bubbles.

• The heat-stable catalase test is based on the ability of the catalase enzyme to remain active after heating (i.e., it is a measure of the enzyme’s heat stability). When heated to 68° C for 20 minutes, the catalase of M. tuberculosis, M. bovis, M. gastri, and M. haemophilum becomes inactivated.

Tween 80 Hydrolysis.

The commonly nonpathogenic, slow-growing scotochromogens and nonphotochromogens produce a lipase that can hydrolyze Tween 80 (the detergent polyoxyethylene sorbitan monooleate) into oleic acid and polyoxyethylated sorbitol, whereas pathogenic species do not. Tween 80 hydrolysis is useful for differentiating species of photochromogens, nonchromogens, and scotochromogens. Because laboratory-prepared media have a very short shelf life, the CDC recommends use of a commercial Tween 80 hydrolysis substrate (Becton-Dickinson, Franklin Lakes, N.J. or Remel Laboratories, Lenexa, Kansas) that is stable for up to 1 year.

Tellurite Reduction.

Some species of mycobacteria reduce potassium tellurite at variable rates. The ability to reduce tellurite in 3 to 4 days distinguishes members of MAC from most other nonchromogenic species. All rapid-growers reduce tellurite in 3 days.

Arylsulfatase.

The enzyme arylsulfatase is present in most mycobacteria. Test conditions can be varied to differentiate different forms of the enzyme. The rate at which this enzyme breaks down phenolphthalein disulfate into phenolphthalein (which forms a red color in the presence of sodium bicarbonate) and other salts helps to differentiate certain strains of mycobacteria. The 3-day test is particularly useful for identifying the potentially pathogenic rapid-growers M. fortuitum and M. chelonae. Slow-growing M. marinum and M. szulgai are positive in the 14-day test (Figure 43-8).

Growth Inhibition by Thiophene-2-Carboxylic Acid Hydrazide (TCH).

This test is used to distinguish M. bovis from M. tuberculosis, because only M. bovis is unable to grow in the presence of 10 mg/mL of TCH.

Other Tests.

Other tests are often performed to make more subtle distinctions between species (see Table 43-11). However, performing all the procedures necessary for definitive identification of mycobacteria is not cost-effective for routine clinical microbiology laboratories; therefore, specimens that require further testing can be forwarded to regional laboratories.

Direct Versus Indirect Susceptibility Testing

Susceptibility tests may be performed by either the direct or indirect method. The direct method uses as the inoculum a smear-positive concentrate containing more than 50 acid-fast bacilli per 100 oil immersion fields; the indirect method uses a culture as the inoculum source. Although direct testing provides more rapid results, it is less standardized, and contamination may occur.

Conventional Methods

The development of primary drug resistance in tuberculosis represents an increase in the proportion of resistant organisms. This increase in resistant organisms results from a spontaneous mutation and subsequent selection to predominance of these drug-resistant mutants by the action of a single or ineffective drug therapy. A poor clinical outcome is predicted with an agent when more than 1% of bacilli in the test population are resistant. If an isolate is reported as resistant to a drug, treatment failure is likely if this drug is used for therapy.

Drug resistance is defined for M. tuberculosis complex in terms of the critical concentration of the drug. The critical concentration of a drug is the amount of drug required to prevent growth above the 1% threshold of the test population of tubercle bacilli.

Four general methods are used throughout the world to determine the susceptibility of M. tuberculosis isolates to various antituberculous agents (Table 43-12). Initial isolates of M. tuberculosis are tested against five antimicrobials, which are referred to as primary drugs (Box 43-4.) If resistance to any of the primary drugs is detected, a second battery of agents is tested (Box 43-4).

New Approaches

Several technologies recently introduced show promise of being faster, more reliable, and/or easier to perform than most conventional methods of susceptibility testing. For example, mutations leading to rifampin resistance have been detected using molecular methods. One molecular method, the line probe assay (INNO-LiPA Rif TB; Innogenetics, Ghent, Belgium), is a commercially available, reverse hybridization–based probe assay for rapid detection of rifampin mutations leading to rifampin resistance in M. tuberculosis. Many different genotypic assays are currently available for drug susceptibility testing. Most are based on PCR amplification of a specific region of an M. tuberculosis gene, followed by analysis of the amplicon for specific mutations associated with resistance to a particular drug. The presence or absence of mutations can then be detected by several methods, such as automated sequencing.

As previously mentioned, high-density DNA probe assays (see Chapter 8) have been used to detect rifampin resistance and to identify mycobacterial species identification.

An innovative approach used by Jacobs et al.³ to perform susceptibility testing involved the use of a luciferase-reporter mycobacteriophage (bacterial viruses). The basis for this assay is simple: viable mycobacteria can become infected with and replicate the mycobacteriophage; dead tubercle bacilli cannot. The mycobacteriophage was constructed to have the firefly luciferase gene next to a mycobacterial promoter; therefore, the presence and growth of the mycobacteriophage is detected by chemiluminescence. In brief, the isolate of M. tuberculosis to be tested is grown in the presence and absence of drug, and the specially constructed mycobacteriophage is added. After infection, luciferin, a substrate of luciferase, is added. If organisms are viable (i.e., thereby allowing infection of the bacteriophage and subsequent transcription and translation of the luciferase gene), the luciferin is broken down and light is emitted that can be measured; the amount of light emitted is directly proportional to the number of viable M. tuberculosis organisms. Therefore, if an organism is resistant to the drug, light is emitted; organisms susceptible to the drug do not emit light. Another commercially available assay that uses mycobacteriophages is the FAST Plaque TB–RIF test (Bio Tec Laboratories, Ipswich, UK).

Susceptibility testing should be repeated if the patient remains culture positive after 3 months following appropriate therapy or fails to respond clinically to therapy.

Therapy

Therapy directed against M. tuberculosis depends on the susceptibility of the isolate to various antimicrobial agents. To prevent the selection of resistant mutants, treatment of tuberculosis requires four drugs: isoniazid, rifampin, ethambutol, and pyrazinamide. Initial therapy includes all four drugs for 8 weeks. However, if drug susceptibility is determined for isoniazid, rifampin, and pyrazinamide, ethambutol may be discontinued. This is the preferred therapy for initial treatment, followed by isoniazid and rifampin for an additional 18 weeks. The most common two-drug regimen is isoniazid (INH, also known as isonicotinylhydrazine) and rifampin. The combination is administered for 9 months in cases of uncomplicated tuberculosis; if pyrazinamide is added to this regimen during the first 2 months, the total duration of therapy can be shortened to 6 months. Ethambutol may also be added to the regimen. INH prophylaxis is recommended for individuals with a recent skin test conversion who are disease free.