Wallach's Interpretation of Diagnostic Tests: Pathways to Arriving at a Clinical Diagnosis (584 page)

   Cultures should include liquid media and at least one type of solid media. Growth is more rapid in liquid media. Solid media may be more sensitive for isolation of mycobacteria and can also provide information about the quantity of growth, colony morphology, and purity of culture.

   Broth systems have been used to develop automated systems for incubation and detection of growth. Automated systems have decreased turnaround time for positive cultures and are less labor intensive than traditional culture methods.

Direct detection (AFB smear)
:

   The acid-fast smear is used for direct detection of mycobacteria in clinical specimens; Gram stains are not reliable for detection. AFB should be quantified (e.g., 1+ to 4+) for positive smears.

   The sensitivity of AFB smears for detection of tuberculosis is variable (20– 80%), depending on factors like type of disease, specimen quality, laboratory procedures, and experience; overall, at least one AFB smear is positive in 60–80% of patients with active tuberculosis and positive cultures for
Mtb
.

   Sensitivity of the AFB smear is directly related to the organism burden in the sample; detection is improved by evaluation of multiple specimens, examination of sputum samples of >5-mL volume, decontamination and concentration, use of fluorochrome staining, and following standardized methods for smear examination.

   The predictive value of a positive AFB smear for tuberculosis is >90%.

Molecular testing

   Several FDA-approved nucleic acid amplification assays are available for the direct detection of
Mtb
in clinical specimens. Good performance depends on strict adherence to the manufacturer’s instructions.

   The sensitivity for detection is intermediate between AFB smear and culture; assays are useful to provide presumptive diagnosis of
Mtb
infection in patients with positive AFB smears (sensitivity: 40–77% in smear-negative patients and >95% in smear-positive patients).

Testing should only be performed on patients with a high clinical suspicion for tuberculosis. Though the specificity is high (>95%), the clinical utility may be unacceptably low in low-prevalence populations because of falsepositive test results.

   Nonamplified rRNA probes are available for preliminary identification of some mycobacterial species from positive cultures, including
M. tuberculosis
complex,
M. avium
and
intracellulare
,
M. gordonae,
and
M. kansasii
. Note: The
M. tuberculosis
complex includes
Mtb
,
M. bovis
,
M. africanum
,
M. microti
, and several other related species.

Susceptibility testing
: Should be performed on all initial Mtb isolates and repeated if cultures remain positive after 3 months of appropriate treatment. Susceptibility testing for second-line agents should be performed on rifampin-resistant isolates, on isolates resistant to any two other primary drugs, or for patients in whom a second-line agent will be used for treatment.

   
Method
: The agar proportion method, using organisms isolated in culture, is commonly used for susceptibility testing. Standardized inocula of the clinical isolate are inoculated onto Middlebrook plates containing a specific critical concentration of the drug tested, as well as drug-free control media. Antibiotics for which there is <99% reduction in organisms, compared to growth on the control media, are unlikely to be clinically effective. Susceptibility test methods adapted to use of liquid media have been developed using automated or manual methods. Testing methods have also been described for direct preparation of inocula from smear-positive specimens.

   
Primary panel
: Isoniazid (INH), rifampin (RMP), ethambutol (EMB), and pyrazinamide (PZA).