Frequently asked questions

The near point-of-care nucleic acid amplification tests (NPOC-NAATs) are molecular tests for TB diagnosis that can produce results either from swabbed sputum or tongue swabs in less than one hour using instruments that can be battery operated and require no specialized infrastructure for use or storage, and they can be performed by health care workers with basic technical skills (e.g. basic pipetting) because they do not require precision.

WHO has highlighted three priority use cases for NPOC-NAAT introduction that prioritize opportunities to provide testing where existing services (i.e., low- or moderate-complexity networks) are insufficient or do not reach.

1. Replacing smear microscopy service use for initial TB diagnosis, while retaining microscopy use for treatment monitoring, where needed. WHO recommends replacing smear microscopy used to diagnose TB with molecular WRDs, which offer higher accuracy than smear microscopy, are easier to do, and support testing when sputum is not obtainable by using other sample types, such as tongue swabs.
2. Introducing molecular testing services at intermediate & decentralized laboratories where other molecular tests are needed and access to testing cannot otherwise be achieved in the immediate or short-term. For this approach, use of NPOC-NAATs should be reviewed when other, more sensitive mWRDs with drug resistance testing become available (i.e., LC-aNAATs or MC-aNAATs) to support closer-to-client diagnosis of TB and drug resistance.  
3. Providing testing services in hard-to-reach areas, such as peripheral primary healthcare & community settings where existing laboratory networks may not provide initial diagnostic services within the WHO-recommended 48-hour turnaround time, but referral for follow-on testing and linkage to care can be assured.

As with any WRD, the placement of NPOC-NAATs should be guided by national strategic plans and priorities, local epidemiology, infrastructure and staff availability. NPOC-NAATs do not yet include drug susceptibility testing (DST). In settings with a high prevalence of rifampicin (RIF)-resistant TB it may be more efficient to place or link facilities to existing LC-aNAAT/ MC-aNAAT sites to ensure timely drug resistance results are available for care and avoid having to refer and retest all TB positive samples at additional cost. WHO does not recommend prioritizing replacement of existing molecular WRD testing capacity with NPOC-NAATs where such capacity is already available.

No. WHO does not recommend replacing existing molecular WHO-recommended rapid diagnostic tests (WRDs), such as LC-aNAATs or MC-aNAATs, with NPOC-NAATs. NPOC-NAATs are intended to complement existing diagnostic networks, particularly in decentralized or hard-to-reach settings and in settings currently relying on smear microscopy. NPOC-NAATs allow for testing programs to offer initial molecular testing for TB where it may otherwise not be available. However, these tests are less sensitive than LC-aNAATs and MC-aNAATs, do not detect TB as well in people with low and variable amounts of MTBC in their samples as LC-aNAATs and MC-aNAATs, and do not currently provide drug resistance results.

NPOC-NAATs are recommended to diagnose TB in children only using swabbed sputum (expectorated or induced). This recommendation was based on extrapolation from data from adults, and limited data from children (no data were available to estimate sensitivity for children using NPOC-NAATs on sputum).

The recommendation does not apply to the use of tongue swabs or to paediatric samples other than sputum (i.e. stool, bronchoalveolar lavage, gastric aspirates or nasopharyngeal aspirates due to a lack of evidence on the impact and accuracy of these samples with near point-of-care tests. Studies are ongoing to generate this evidence and will be reviewed by WHO, once available.

Yes, NPOC-NAATs are recommended for adults and adolescents living with HIV using both swabbed sputum and tongue swabs. However, evidence for this group was limited, and no evidence was available for people with advanced HIV disease or severe illness. The summary sensitivity (95% CI) in people living with HIV was 79.2% (68.7 to 86.9) and the summary specificity was 96.7% (94.6 to 98.1) (1). Wherever available, concurrent testing with LC-aNAATs on a respiratory sample and LF-LAM is the preferred testing strategy for people living with HIV because this diagnostic strategy has the best performance in this population group. If a sputum specimen cannot be obtained, a tongue swab may be used as part of concurrent testing for this population group alongside LF-LAM on urine.

No, biosafety cabinets are not required. Use of NPOC-NAATs, including sputum and swab-based testing, is equal in biosafety risk to smear microscopy.

Yes, there is a risk of contamination. NPOC-NAATs are rather sensitive for detection of M. tuberculosis DNA, and even small lapses in cleaning between samples can lead to cross-contamination between samples, which can produce false-positive results.

No, sputum pooling as a diagnostic strategy to reduce costs during periods of resource constraint was not assessed for NPOC-NAATs. There is not currently evidence on how pooling would affect clinical outcomes, diagnostic accuracy, cost effectiveness, or end user values associated with these technologies. Pooling is currently recommended only for specific LC-aNAAT technologies (Xpert Ultra) within existing laboratory testing networks. As NPOC-NAATs were found to have slightly reduced sensitivity compared to LC-aNAATs and may not detect low levels of bacteria in samples as well as LC-aNAATs, this evidence requires assessment.

No, current WHO-recommended NPOC-NAATs do not detect resistance to rifampicin (RIF) or other anti-TB drugs. Therefore, individuals with a positive TB test result should undergo additional testing using a WHO-recommended molecular assay capable of detecting drug resistance. As sputum is the preferred sample type for molecular pulmonary TB testing, all efforts should be made to collect and refer sputum for this purpose. If sputum is not available, tongue swabs may be used for both TB and rifampicin resistance detection on LC-aNAATs compatible with this sample type.

Advantages:

• Can be battery-operated and done in basic peripheral laboratories (e.g. those that perform diagnostic smear microscopy) and in health clinics, mobile units or community sites that do not have laboratories but have access to follow-on testing and treatment services;
• Do not require temperature-control environments for instruments or reagents or dust protection for instruments;
• Can be performed by health care workers with basic technical skills (e.g. basic pipetting) because they do not require precision.
• Produce pre-test to post-test results in <40 minutes from either swabbed sputum (preferred) or  tongue swabs (when sputum cannot be produced)
• Can be used with tongue swabs, which expands testing access to people who cannot provide sputum;
• Instruments can be used to test for multiple diseases (tests are currently available for influenza, RSV, COVID-19, and HPV, with others in development).
• The first-in-class test and the instruments are available at lower cost than other molecular WHO-recommended diagnostics for the detection of TB which may support expansion of access in decentralized settings. 

Disadvantages:

• These technologies do not detect rifampicin resistance;
• While judged to be very accurate, lower sensitivity than LC-aNAATs on sputum (85% vs 90%) (1), which means that certain TB patients with low bacterial burden/paucibacillary disease can be missed.
• The test performance on different sample types is also variable, with tongue swabs less sensitive than sputum (76% vs. 85%) among mostly sputum-producers (1).
• Enhanced personal protective equipment (PPE) required for healthcare worker collection of tongue swabs (e.g., eye protection, N95 mask, gloves, coat or overalls), more similar to COVID-19 due to the proximity required for sample collection;
• The test is instrument-based; testing requires two instruments and battery charging with charging and replacement mechanisms for testing sites that do not rely on stable power.
• Processing more than one sample at a time, and sample transfer between lysis and testing, can lead to contamination (i.e., false positive results for serially tested samples). Contamination can occur due to accidental spills or imprecise drops of samples or sample solutions onto tester gloves, instruments or work surfaces. Disinfectants, spill kits and best testing practices must be implemented to reduce risk of contamination.

A tongue swab is a biological sample obtained by a simple sampling method where a sterile swab is used to collect material from the surface of the tongue to test for infections (1).

Yes, duration of tongue swab collection is important; evidence shows that sensitivity for diagnosing TB decreases when swabbing is done for less than 30 seconds. Testing sites should have timers available and testers trained on their use and calibration.

Tongue swabs are recommended for initial detection of TB among adults and adolescents living with HIV both with NPOC-NAATs and LC-aNAATs. However, sensitivity of tongue swab testing is reduced compared to sputum testing, so all efforts should be taken to collect and test sputum, whenever possible. Irrespective of whether sputum or tongue swabs are used for initial molecular testing, point-of-care LF-LAM testing on urine should be done concurrently among people living with HIV to maximize opportunities for rapid diagnosis.

Tongue swab collection should incorporate appropriate infection control and biosafety measures, including sampling in a well-ventilated space (2). Due to the close proximity required for healthcare workers to collect tongue swabs from people with presumed TB, enhanced PPE should be used by  healthcare workers during sample collection (e.g. eye protection such as goggles or face shields, N95 masks, gloves, coats and overalls).

Two classes of diagnostic tests for the initial detection of TB are recommended for use with tongue swabs:

1. Near point-of-care nucleic acid amplification tests (NPOC-NAATs)
2. Low-complexity automated nucleic acid amplification tests (LC-aNAATs)

No, presently tongue swabs are not currently recommended for use in children with NPOC-NAATs (no evidence) or with LC-aNAATs (limited evidence showing very low sensitivity). Multiple studies are assessing the performance of tongue swabs in children and other populations at increased risk of TB; WHO will review the evidence once it becomes available.

No, sputum is the preferred sample type for diagnosis of tuberculosis for all WHO recommended initial tests, because it provides a more accurate result for TB and rifampicin resistance detection and can be used to detect lower levels of TB in the sample. However, if sputum cannot be obtained from a person with presumed TB or that screens positive for TB, a tongue swab is recommended for use with NPOC-NAATs and/or LC-aNAATs to provide an opportunity for bacteriological confirmation of disease where no other option may exist.

Advantages:
1. Enables testing of people with signs and symptoms of TB, or who screen positive for TB, who cannot produce sputum.
2. Allows use of existing tests, instruments, and testing networks that are immediately available (no need to wait for NPOC-NAAT regulatory approval and uptake).
3. Allows for simultaneous detection of TB and rifampicin resistance using Xpert MTB/RIF Ultra.

Disadvantages:
1. Performance of LC-aNAATs on tongue swabs is less sensitive than LC-aNAATs on sputum (71% vs 90%) (3,4).
2. Enhanced personal protective equipment (PPE) required for healthcare worker collection of tongue swabs (e.g., eye protection, N95 mask, gloves, coat or overalls), more similar to COVID-19 due to the proximity required for sample collection.

Tongue swabs are recommended for use with two WRDs classes (NPOC-NAATs and LC-aNAATs) as part of initial diagnostic tests for TB detection, in situations when sputum cannot be obtained. National TB diagnostic algorithms should emphasize sputum collection, but may recommend tongue swab collection from adults and adolescents when sputum cannot be obtained. Wherever available, concurrent testing with LC-aNAATs on a respiratory sample and LF-LAM is the preferred testing strategy for people living with HIV. If a sputum specimen cannot be obtained, a tongue swab may be used as part of concurrent testing for this population group alongside LF-LAM on urine.

Sputum pooling is a new diagnostic strategy for TB LC-aNAAT testing in which up to four individual samples are mixed and tested together as part of a single ‘pooled’ sample. The strategy can improve testing efficiency and reduce costs in resource-constrained environments. This strategy has been used for other infectious diseases (e.g. blood bank screening for infectious hepatitis) and can lead to reductions in test, reagent and commodity costs, machine time and laboratory workload.

Testing of pooled sputum for TB is a two-step process. For LC-aNAAT testing, each sputum sample that is identified for pooling receives an unique identifier, so that it is possible to go back and trace the single samples and the patients that have provided them, when the pooled test is positive. Next, each sample is individually treated with sample processing reagents according to manufacturer instructions. A portion of each liquefied sputum sample is then added, one at a time to the pool, labelled with a unique identifier for their pool, and the pooled sample is tested. All of the individual and pooled sputum samples are stored under manufacturer-recommended conditions until test results are available.

If a pooled sample tests negative, then no further testing is performed because it is presumed that all of the individual samples that contributed to the pool are negative for TB detection. If a pooled sample tests positive, then the individual samples that contributed to the pool must be re-accessed and tested individually to identify which of them is positive. One or more of the individual samples in the pool may be positive for TB. 

Even though selected LC-aNAATs (Xpert Ultra) technologies can provide results for rifampicin resistance in parallel to results for TB detection, results for rifampicin resistance from pooled testing are not considered to be reliable and are therefore not suitable for clinical management. Individual samples of pools with detected resistance to rifampicin should be re-tested individually to confirm the presence of resistance to rifampicin. 

Pooling of sputa for the detection of tuberculosis does not apply to people living with HIV (limited evidence), children (insufficient evidence) or individuals at increased risk of drug resistance owing to the potential for reduced equity and increased harm. These and other priority populations may be prioritized for individual testing according to local risk groups.

Advantages:
• Pooling can reduce costs down to as low as 3USD per test, and increase cost-effectiveness for TB testing programs.
• In settings with ≤24% TB positivity rates, pooling can reduce the number of tests needed for the same number of people – enabling more individuals to be tested with fewer tests and less instrument run time (1).
• Test results may be returned faster from high-volume TB testing sites, as pooling allows for testing of up to four sputa in one run, reducing the number of tests and modules required in a given amount of time.  
• Pooled testing can be performed using existing GeneXpert testing services which are immediately available.

Disadvantages:
• Evidence is currently only available for one technology (i.e. Xpert MTB/RIF Ultra).
• This approach is not recommended for vulnerable groups (children, PLHIV, people with risk of DR-TB, other setting-specific populations).
• The cost saving of pooling is dependent on TB prevalence and test positivity rate; as prevalence and positivity rate increase, unit cost savings decrease due to the higher number of required re-tests.
• This strategy increases complexity of sample management for testing, including sample management for testing, sample processing with increased risk of contamination, appropriate storage of accurately-grouped samples until pooled results are available, and recording and reporting of results for pools and individual samples.
• Sensitivity for this method was further reduced when sputum was collected in community-based case-finding (sensitivity range 25-96%, 4 studies) vs facility-based case-finding (sensitivity range 84-100%, 7 studies). The reasons for this variability were not directly assessed but may include bacterial load or other factors (1).