TB Culture and Drug Susceptibility Testing

Mycobacterial culture is the reference method for diagnosing tuberculosis and for obtaining an isolate for species identification and drug susceptibility testing (DST).

Specimens are decontaminated and concentrated (N-acetyl-L-cysteine with sodium hydroxide), then inoculated into liquid medium in an automated system (Mycobacteria Growth Indicator Tube, MGIT 960, which detects growth by oxygen consumption) and onto solid medium (Lowenstein-Jensen or Middlebrook 7H10 and 7H11).

Liquid culture is more sensitive and faster (positive in a median of 10 to 14 days, up to 6 weeks) than solid culture (3 to 8 weeks). Growth is confirmed as acid fast, and identified as Mycobacterium tuberculosis complex or non-tuberculous mycobacteria by immunochromatographic antigen tests (MPT64), line probe assays, sequencing or MALDI-TOF.

Culture detects as few as 10 to 100 viable bacilli in each mL, compared with 5000 to 10 000 for smear, and is positive in 80 to 85% of pulmonary tuberculosis and in about 30 to 50% of smear negative cases. Drug susceptibility testing identifies resistance, and drug resistant tuberculosis affects about 450 000 people each year (rifampicin resistant or multidrug resistant).

Phenotypic DST in liquid medium with critical concentrations remains the reference for most drugs, with genotypic methods (molecular and whole genome sequencing) increasingly used.

Indications. Every patient with suspected or confirmed tuberculosis should have at least one specimen cultured.

Smear negative suspected pulmonary tuberculosis, extrapulmonary disease (pleural fluid, CSF, lymph node, pus, urine, bone marrow, tissue, gastric aspirate), paediatric tuberculosis, HIV-associated tuberculosis, suspected non-tuberculous mycobacterial infection, treatment failure or persistent smear positivity at 2 months or later, relapse, contacts of patients with drug resistant tuberculosis, and monitoring of drug resistant tuberculosis (monthly culture).

Specimen. Sputum (two to three early morning samples of 3 to 5 mL), induced sputum, bronchoalveolar lavage, gastric aspirate (3 consecutive mornings in children, neutralised with sodium bicarbonate), CSF (at least 5 to 10 mL), pleural fluid (50 mL or more) and tissue biopsy (in sterile saline, not formalin), urine (early morning, three samples), blood (mycobacterial blood culture in HIV), pus and aspirates.

Stored at 2 to 8 °C and delivered within 48 hours. Containers must be sterile, leak proof and labelled.

Reference values. No growth after 6 weeks in liquid medium and 8 weeks on solid medium is reported as negative. Positive cultures are reported with the identification and, where tested, susceptibility (susceptible, resistant) for each drug.

Interpretation.

• Culture positive for Mycobacterium tuberculosis complex confirms active tuberculosis (with rare exception of laboratory cross-contamination).
• Culture positive for non-tuberculous mycobacteria: significance depends on repeated isolation, clinical and radiological features (pulmonary disease criteria need at least two positive sputum cultures, or one from bronchial wash or lavage, or a biopsy with granuloma and a positive culture). A single positive from a non-sterile site may be contamination.
• Culture negative with strong clinical suspicion: tuberculosis is not excluded (paucibacillary disease), and treatment decisions combine radiology, histology (caseating granuloma), molecular tests and response to treatment.
• Phenotypic DST: resistance to isoniazid and rifampicin defines multidrug resistant tuberculosis (MDR-TB). MDR plus resistance to a fluoroquinolone defines pre-extensively drug resistant disease, and additional resistance to bedaquiline or linezolid defines extensively drug resistant disease. Rifampicin resistance alone is treated as MDR.
• Critical concentrations (MGIT): rifampicin 1.0 micrograms/mL, isoniazid 0.1 micrograms/mL (high level 0.4), ethambutol 5.0, pyrazinamide 100 (pH 5.9), levofloxacin 1.0, moxifloxacin 0.25, bedaquiline 1.0, linezolid 1.0. Results should be reported to the clinician within 1 to 2 weeks of a positive culture.
• Culture conversion (negative cultures) at 2 months is a marker of early treatment response (about 85% of drug susceptible patients). Positive culture after 4 months defines failure, and a new DST is required.
• Contamination of cultures (bacterial or fungal overgrowth) occurs in 3 to 8% of liquid cultures, and the sample is repeated.

Critical values and findings needing immediate action.

• Rifampicin or multidrug resistance on culture or molecular testing: stop the usual first line regimen, and refer to a drug resistant tuberculosis service for an all-oral regimen with drugs chosen by susceptibility results.
• Growth of Mycobacterium tuberculosis in CSF, blood or in an immunocompromised child: severe disseminated tuberculosis, urgent treatment and public health notification.
• Positive culture in a healthcare worker, child contact or pregnant woman: urgent treatment and contact tracing.
• Resistance to fluoroquinolones in a rifampicin resistant isolate: specialist regimen design (bedaquiline, pretomanid, linezolid, delamanid).
• Positive culture during treatment after 4 to 5 months of therapy: treatment failure, adherence assessment, repeat DST and sequencing.

Interferences and pitfalls.

• Culture is slow (weeks), so treatment starts on clinical grounds and molecular testing, not waiting for culture. DST results arrive 2 to 6 weeks after sampling.
• Sensitivity is reduced by prior anti-tuberculosis therapy, delayed transport, poor sample quality, over decontamination and paucibacillary disease.
• Critical concentrations for some drugs (ethambutol, pyrazinamide, ethionamide, cycloserine) give poor reproducibility, and borderline results need confirmation by genotypic testing.
• Heteroresistance (mixed susceptible and resistant populations) may be missed by phenotypic testing or low frequency mutations by molecular tests.
• Laboratory cross-contamination causes false positive cultures (especially a single positive culture with scanty or negative smear), and genotyping can identify it.
• Culture handling needs a biosafety level 3 facility for drug resistant and culture work, and many settings lack it.
• Bacterial and fungal contamination reduces the usable yield, and nasal and oral colonisers can cause overgrowth.
• Non-tuberculous mycobacterial DST follows different methods and breakpoints (macrolides for Mycobacterium avium complex, amikacin, and others).
• Interpretation of cultures in patients on treatment requires differentiation of viable bacilli from dead organisms detected by smear.

Special groups.

• Children: difficult sampling, so gastric aspirates, induced sputum and stool molecular tests are used, and culture confirmation is available in 30 to 40% of children with presumptive tuberculosis.
• Pregnancy: first line treatment is safe, and for drug resistant tuberculosis specialist advice on regimen is needed.
• HIV: more extrapulmonary disease, mycobacteraemia, and sputum negative disease, so blood and urine culture are used.
• Older adults: lower yield, atypical presentation, and non-tuberculous mycobacteria are more frequent.
• Regions: culture and DST capacity is limited in many low income countries, and sample transport networks and referral laboratories are used. Whole genome sequencing for resistance prediction is expanding.

Follow up tests. Xpert MTB/RIF or Ultra, line probe assay, targeted or whole genome sequencing for resistance mutations, smear microscopy, species identification (MPT64, line probe, sequencing), HIV test, chest radiograph or CT, liver function tests, hearing and renal monitoring with injectable drugs, ECG (QT interval) with bedaquiline, delamanid and fluoroquinolones, and monthly culture during drug resistant tuberculosis treatment.