India remains the country with the world’s highest TB burden, with an estimated incidence of 2,950,000 (2,510,000–3,440,000) and with notification of 2.14 million TB cases in 2021; this is 18% higher than the incidence in 2020 and also represents 27% of global TB case notifications in 2021 (1). However, there was a very large gap between the number of people who fell ill with TB in 2021 and the number newly diagnosed and reported, as compared to 2019. This global gap may be attributed to both under-diagnosis and under-reporting of TB cases. To enable rapid TB testing, the WHO has recommended low- and moderate-complexity nucleic acid amplification technology (NAAT) for use in the initial diagnostic test. Among the forms of low-complexity NAAT testing available for use in peripheral settings, the WHO has recommended the use of Xpert MTB/RIF, Xpert MTB Ultra, Truenat MTB, MTB plus, MTB-RIF Dx, and TB LAMP assays (2). Of the 49 high-burden countries globally, 26 countries have reported the use of molecular tests as an initial diagnostic test for more than half of their notified cases (1). Following the WHO recommendation, India was the first country to implement Truenat MTB-RIF as an upfront molecular test, as early as 2020, under the National TB Elimination Program (NTEP) (3, 4).

Truenat MTB-RIF assay is a novel TB diagnostic tool developed by Molbio Diagnostics Pvt. Ltd., Goa, India. It is a real-time polymerase chain reaction (RT-PCR) chip-based assay that can detect Mycobacterium tuberculosis (Mtb) and rifampicin (RIF) drug resistance. With the combined advantages of affordability, simplicity in operations, diagnostic sensitivity, and portability, this micro-PCR device represents a strong candidate for wide-scale use in resource-limited settings (5).

Truenat MTB-RIF was implemented under the restrictive circumstances of COVID-19 in 2020, when TB testing was enormously reduced due to countrywide re-routing of Truenat machines for COVID-19 testing. By the end of 2021, over 1,972 Truenat machines were deployed at the district and subdistrict level under the NTEP in order to fast-track the upfront molecular testing of presumptive TB patients and identify rifampicin drug resistance in Mtb-positive specimens. However, the positioning of this rapid molecular TB diagnostic tool across peripheral laboratories under the NTEP in India was challenged by the rates of inconclusive Truenat results (“invalid,” “indeterminate,” or “error”). A multicenter validation study conducted previously in four reference laboratories under the NTEP showed operational advantages for the use of Truenat as an upfront molecular test (6). The diagnostic accuracy of Truenat in primary health settings and reference laboratories has been previously evaluated and found to be sufficient in a multicenter study across four different countries (7). Both studies were conducted in fewer testing centers with a controlled set-up, a limited sample size, and appropriate training of the laboratory technicians. In contrast, the present study aimed to assess the rate of inconclusive Truenat results in real-world situations and to determine their root causes. This knowledge is needed in order to optimize the testing performance of Truenat MTB-RIF, to ensure timely diagnosis, and thereby to reduce the magnitude of undiagnosed TB cases, not only in India but also in other countries that have rolled out Truenat as a molecular point-of-care tool to strengthen the TB diagnostic care cascade in national TB programs.

The analysis of 327,649 presumptive TB patients tested in a 3-month period at 1,690 Truenat sites distributed across 89% of the states and UTs of India (n = 32) indicated that 22.1% of presumptive TB cases were Mtb-positive. This rate was higher than the 14.1% reported by Adam Penn-Nicholson et al. in a multi-country study (7) or the 13.0% reported by Abyot Meaza et al. in Ethiopia (8). The reason for the higher rate of Mtb-positives may be the testing of samples from smear-positive TB patients for universal drug susceptibility testing (UDST) in addition to presumptive TB patients at a small number of the sites. In India, a total of 2,197,757 Truenat MTB tests were performed in 2021, with a positivity rate of 20.5% (9). Our study indicated an MTB positivity rate of 22.1% during the period April to June 2022 in Truenat MTB-RIF testing.

In programmatic settings in India, a second sample from patients determined to be positive for Mtb should be sent to a culture and drug susceptibility testing (DST) facility for first−/second-line probe assay (FL/SL LPA), followed by liquid culture DST for all drug-resistant samples detected by LPA (10). In 2021, 29,84,636 presumptive TB patients were tested via NAAT, and Mtb was detected in 8,73,725. However, only 3,28,715 samples were tested via FL-LPA, and 46,733 resistant TB cases were identified. SL-DST was performed for 14,886 samples, but there are no reports on the numbers of samples tested by DST facilities for newer drugs (9). These numbers indicate a gap between initial and follow-on resistance testing and reinforce the need for efficient upfront molecular testing. The Truenat testing system has great aspirations to fill in this gap in the diagnostic pipeline. Under the DR-TB regimen, all patients except those whose samples indicate isoniazid resistance are started on an MDR/RR-TB regimen, and early identification of rifampicin resistance status could help with initiation of this treatment regimen.

Hands-on training is important for Truenat testing as, unlike GeneXpert, the process involves additional manual steps for DNA extraction and requires the handling of pipettes. Under a program fo POC molecular diagnosis, such as Truenat, it becomes highly critical that the lab technicians are well trained in the various methods and processes involved in the technique, such as micropipette handling and knowledge of workstation cleanliness, disinfection techniques, and proper operation of instruments. Even subtle factors in handling micropipettes, such as immersion angle, immersion depth, ergonomic practices, maintenance, and periodic servicing, are highly critical for good laboratory practice, which would help to consistently produce a higher rate of concordant molecular diagnosis of TB (11). Continuous monitoring of the competency of LTs by STLS/supervisory staff would help to resolve issues at the earliest possible stage. Although training of trainers was conducted by the NTEP in March 2020, cascade training of IRLs and district staff could not be carried out due to the COVID-19 pandemic. Virtual training sessions were conducted for the staff, which were not as effective as in-person, hands-on training. Considering this gap due to unforeseen situations, it has recently been suggested by the NTEP that all the IRLs conduct in-person training of staff.

Inconclusive results on the Truenat MTB-RIF occur for multifactorial reasons, all of which need to be addressed, and gaps between policies and testing efficiency must be identified (12, 13). In this study, the rates of invalid Truenat MTB/RIF results were lower (5.2%) compared to the 11.6% reported in Ethiopia (8), while the rate of inconclusive results (6.7%) corroborated those of a previous study in India (6). Despite the observed low proportions of inconclusive RR results (16.9%), these unsuccessful outcomes could lead to meaningful missed opportunities in DR-TB case detection. The lack of retesting of inconclusive RR results indicates the potential negative impact that the inconclusive Truenat results may have in DR-TB case detection. The study identified a testing gap of 41.1 and 78.2% for MTB and RIF testing, indicating a lack of knowledge of the current Truenat testing algorithm in India recommending retesting in the case of inconclusive results. The study indicates a need to draw up a plan for rigorous training emphasizing iteration of Truenat MTB-RIF testing to obtain a conclusive result. Samples eliciting inconclusive Truenat MTB-RIF results are recommended to be tested at the same testing site using the same DNA eluate, and if the outcome is not resolved, then a fresh specimen can be used (5, 10). A multicenter study conducted by FIND (the Foundation for Innovative New Diagnostics) showed that retesting of samples reduces inconclusive results from 6.2 to 1.7% for the Truenat MTB test (7). The performance indicator recommended for the rate of inconclusive results is <3% for MTB testing (5). With rigorous training, competency assessment, and quality monitoring by the Central TB Division (Ministry of Health and Family Welfare), with the respective NRL and IRL, the nation could achieve the expected level of technical proficiency. Additionally, the rate of indeterminate RIF results has been reported to be high for Truenat MTB testing, even after retesting, particularly when the bacillary load was low (7). The quality of the sample is also critical for proper diagnosis when using extrapulmonary specimens. Paucibacillary status and the presence of potential inhibitors of PCR in non-sputum samples pose additional challenges for the diagnosis of TB using molecular techniques, such as by the Truenat method (14, 15). Previous studies on Xpert MTB/RIF have also indicated similar issues with sample processing and operator error, indicating technical, machine maintenance, and operational issues leading to higher rates of inconclusive results (16, 17).

The MTB test results provide a cycle threshold (cT) value for the target gene and an estimated bacterial load in the form of colony-forming units per ml (CFU/ml) (5). Based on our study, we recommend the inclusion of cT and CFU/ml as additional information in the performance indicators for Truenat MTB testing. Low bacillary load samples and their corresponding RIF status could be validated, and this would improve the overall implementation and effectiveness of the Truenat MTB/RIF test.

The root cause analysis visits conducted at 12 selected sites showed that the main causes of inconclusive results included faults related to machine maintenance, inadequate sample quality, and inadequate quality control. The Guidelines for Truenat testing recommend performing a quality control test with negative and positive control samples routinely and on occasions when the storage temperature of the Truenat chips falls outside the recommended temperature range of 2–30°C (5). These issues need to be addressed appropriately to ensure the optimal outcome of Truenat testing at the peripheral level. Effective external quality control and formative supervision could help with this objective.

The main observations of our study include:

However, we also observed that a well-performing site can also perform poorly if a well-trained technician is replaced with an untrained one, and vice versa. Supervision and monitoring of performance with proper recording of the reasons for an invalid or indeterminate test result, including error type, is essential. Reinforcement of retesting should be conducted across various sites to ensure a reduction in inconclusive results.

The limitations of this study are related to missing information from the retrospective data collected from the Truenat sites. The lack of complete data on MTB testing or the mismatched or missing data observed in the case of RIF testing may have led to the exclusion of a well-performing or poorly performing site, causing some degree of bias in our analysis. However, any recall bias during the site visit was mitigated by triangulating data from different sources (Truenat machines, physical laboratory records at the sites, and the Truenat indicators submitted to the CTD through designated reference laboratories under the NTEP). Another limitation was the restriction to 12 site visits from among 81 sites listed initially under various criteria. However, owing to travel restrictions due to COVID, the study personnel limited their visits to sites chosen based on their testing rates, performance, geographical location, climatic conditions, and remoteness to ensure that all factors were covered in the root cause analysis.

The present study has provided the desired knowledge on the magnitude and the potential negative impact of inconclusive Truenat results in DR-TB case detection and performed root cause analysis to identify appropriate solutions to optimize Truenat MTB-RIF testing. Our major findings substantiate the significance of retesting samples that produce inconclusive Truenat results and the value of this method in rapid DR-TB treatment initiation. Our key recommendations for Truenat MTB-RIF testing under the National TB Elimination Program of India, as well as other countries that have deployed Truenat MTB-RIF assay in their diagnostic algorithm, in order to optimize the implementation and outcomes of Truenat assay, include:

Our study showed that well-performing sites with high sample loads had met most of these requirements, which enabled good performance in Truenat MTB-RIF testing. The recommendations highlighted above were consolidated into guidance documents and videos and disseminated nationwide by the CTD, Ministry of Health and Family Welfare, Government of India, in order to improve Truenat MTB-RIF testing. The videos emphasized a set of Good Laboratory Practices (GLPs) and “Do’s and Don’ts” in Truenat MTB-RIF testing (18). Guidance on the roles and responsibilities of the program managers, laboratory supervisory staff, and technical staff to help optimize Truenat testing at NTEP sites was prepared (18). This guidance document included key factors, including programmatic aspects; technical aspects; aspects of recording and reporting; quality assurance; and instrument maintenance for efficient implementation, supervision, and performance of Truenat MTB-RIF assay.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by the ICMR-National Institute for Research in Tuberculosis (NIRT), Chennai and ICF Incorporated LLC Institutional Review Board (IRB). The studies were conducted in accordance with the local legislation and institutional requirements. The human samples used in this study were acquired from a by- product of routine care or industry. Written informed consent for participation was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and institutional requirements.

RG: Conceptualization, Writing – original draft. NK: Conceptualization, Writing – review & editing. HV: Conceptualization, Writing – original draft. PR: Writing – review & editing. JK: Writing – review & editing. CP: Writing – review & editing. SK: Writing – review & editing. SG: Writing – review & editing. HY: Writing – review & editing. MM: Writing – review & editing. SA: Writing – review & editing. RR: Writing – review & editing. UA: Writing – review & editing. SaS: Conceptualization, Supervision, Writing – review & editing. SiS: Conceptualization, Supervision, Writing – review & editing.