Introduction

Tuberculosis (TB) remains a significant global health challenge, with an estimated 10.6 million new cases reported and 1.4 million deaths due to TB in 2021. An estimated 4.2 million people with TB were undiagnosed and unreported, a situation that was complicated by the post-Covid challenges of access to diagnosis for persons affected by TB and linkage to care. Nigeria is one of the countries with high TB, TB/HIV and multi-drug resistant TB (MDR-TB)1.

The diagnostic delays associated with smear microscopy which takes about 24–48 h and solid culture which takes about 4–8 weeks present obstacles to early diagnosis of TB. To achieve the End TB targets, rapid and sensitive molecular diagnostic tools for detection of TB and Rifampicin resistance are required. Over the past decade, the World Health Organization (WHO) has endorsed a new range of diagnostics technologies2. The currently recommended algorithms include the use of rapid molecular-based tests to detect TB and drug-resistant TB. The introduction of the Xpert MTB/RIF assay revolutionized TB diagnosis and gave hope to struggling programmes. This has improved decision making for both drug-susceptible (DS-TB) and Rifampicin-resistant TB (RR-TB) as diagnosis can be made within two hours of sample collection, thereby facilitating prompt initiation of treatment.

Although the use of the GeneXpert assays has been widely scaled up since introduction, the infrastructural requirements, technical challenges and operational costs associated with placement of the system limits the access of these tools to most patients in some settings in high TB burden countries. Ensuring that these molecular tests can be available as point-of-care or near point-of-care increases access to rapid diagnosis and effective treatment and this will be critical to successful tuberculosis (TB) control.

The Truenat assay is a more recently available molecular diagnostic test for TB that is both battery-operated and portable and can be used in the most basic parts of the health system. It has been used primarily in India to support diagnosis of TB. In 2020, following a multicenter evaluation of the Truenat assays conducted by the Foundation for Innovative New Diagnostics (FIND) at seven sites in four countries, the World Health Organization (WHO) recommended the use of the Truenat assay as the initial diagnostic test for TB and Rifampicin resistant TB rather than smear microscopy/culture. The minimal infrastructural and technical skill requirements required for successful operation of the Truenat machine make it an important consideration for deployment to peripheral facilities3,4,5.

Based on the WHO recommendation of Truenat, USAID through the Stop TB Partnership in 2021, procured and donated 38 Truenat machines to Nigeria as part of introducing New Tools Project (iNTP). The iNTP project is the largest multi-country implementation of Truenat instruments which is aimed at providing access to high quality, innovative diagnostic tools for TB among populations in hard-to-reach areas6. The Truenat machines were installed at peripheral facilities that had only microscopy for TB diagnosis and had to transport their sputum samples to GeneXpert laboratories at distant locations. The facilities where they were installed included primary and secondary public and private health facilities. A three-day theory and practical training course was provided for laboratory staff during installation. In addition, a one-day sensitization training was held for health workers selected from within and surrounding facilities.

At facility level, trained laboratory personnel performed the Truenat assay according to the protocol and national standard operating procedures (SOPs). As with any other diagnostic test assay, unsuccessful tests (errors, invalids etc.) are expected for the Truenat assay and these are tracked as part of the National TB Program (NTP) key performance indicators for Laboratory. Based on the SOPs, unsuccessful tests showing errors, invalid or incomplete results were repeated7. The type of error codes displayed by either the Trueprep or Truelab system determines the troubleshooting steps taken to resolve the error. If the errors persist after troubleshooting it is reported to the local Truenat service support agents for further assistance. All errors were recorded in the laboratory registers. Laboratory staff were also trained to perform regular preventive maintenance actions on the machines. The concerns about the unsuccessful test results are the resulting delay in the diagnosis of patients with TB. Considering this is also the first wide-scale roll-out of the Truenat assay outside India, there is yet limited data on the types and frequency of errors encountered in varied field settings where these machines are used. This study therefore sought to determine the frequency and distribution of unsuccessful test results (invalid/error/indeterminate) as well as the possible factors associate with high error rates (> 3%) among Truenat sites in Nigeria.

Methods

Study design

This was a retrospective cross-sectional study that involved analysis of secondary data generated between January 2022 and December 2022 from all thirty-eight (38) Truenat facilities within the 14 implementing states.

Study setting

This study was conducted in the 14 USAID supported TB LON project states implementing the Truenat intervention in Nigeria. The states included Kano, Katsina, Kaduna, Bauchi, Nasarawa and Taraba in the North while Akwa Ibom, Anambra, Cross River, Delta, Rivers, Lagos, Osun and Oyo States are located in the southern part of the country. Three states in southwest Nigeria including Lagos, Osun and Oyo are part of USAID supported TB-LON 3 project implemented by Institute of Human Virology Nigeria, (IHVN) while the remaining 11 states are under the USAID supported TB-LON Regions 1 & 2, implemented by KNCV Tuberculosis Foundation Nigeria.

Study population

Study population included the records of all tests conducted at the Truenat facilities within the time of study as recorded in the Laboratory registers of the study facilities. In addition, information related to the characteristics of the health facilities where the Truenat machine was located, the socio-demographic characteristics, and other related information of the key laboratory staff operating the Truenat assay at the laboratories were obtained. This study was a total sampling of all tests conducted within the study period of January to December 2022.

Study procedure

Research assistants from each of the states implementing Truenat were selected and trained on data collection using the data abstraction tools (Annex I & II). Secondary data from Truenat laboratory registers on number/type of Trueprep errors and number/type of Truelab errors, including unsuccessful test results, were collected by the research assistants for analysis. In addition, socio-demographic data of the Truenat lab staff and specific operational data on preventive maintenance were collected retrospectively from the facility equipment maintenance records. The data collected included a 12-month period data spanning January 2022 -December 2022, at implementing sites across the country.

Data collection

Study tools used in the study included Laboratory registers of the study sites, equipment preventive maintenance records and Truenat Manufacturer’s maintenance guide brochure.

Outcome measures

The outcome measures were the frequency and distribution of unsuccessful test results at Truenat sites and the factors associated with high error rates.

Data management

Data was collected from the laboratory register directly to an excel data template by the data collectors. The data collected at each state level were reviewed and cleaned by the Data Officer before it was shared with the statistician on the central study team for review and further validation before analysis. Data analysis was done using IBM Statistical Product and Service Solutions (SPSS) statistical software version 25. Continuous variables were summarized using mean and standard deviation while categorical variables were presented using frequencies and proportions. Facilities were categorized into those with high error rates (≥ 3%) and those with low error rates (< 3%). In determining the factors associated with high error rates among the Truenat sites, the socio-demographic data from the Truenat lab staff and the specific operational data on preventive maintenance were cross-tabulated with the outcome variable. The level of statistical significance was determined by a p value of < 0.05.

Ethics declarations

Ethical approval for this study was obtained from the National Health Research and Ethics Committee (NHREC) prior to commencement of the study. (Approval number, NHREC/01/01/2007/-12/04//2023). The respondents signed to a written informed consent form and they were assured of confidentiality and anonymity of data. Participation in the study was voluntary and respondents were assured that there would be no victimization of respondents who refused to participate or those who decided to withdraw from the study after giving consent. The study was done in accordance to the Declaration of Helsinki and the National Code for Health Research Ethics in Nigeria.

Results

Table 1 shows the socio-demographic characteristics of the respondents. The mean age of the respondents was 37.2 ± 7.3 years and the highest proportion, 36.8% were less than 35 years. A higher proportion of the respondents, 71.1% were males.

Table 1 Socio-demographic characteristics of the respondents
Full size table

Table 2 shows health facilities and adherence to SOP rules. Majority of the health facilities, 92.1% had the SOP for conducting Truenat visibly pasted on the wall. A higher proportion of the health facilities, 84.2% had preventive maintenance log form visibly kept and recorded.

Table 2 Health facilities and adherence to standard operating procedures (SOP)
Full size table

Table 3 shows the total tests conducted during the study period. A total of 56,799 Trueprep DNA extractions were made and the unsuccessful tests were 1,043 giving an invalid rate of 1.8%. Also a total of 56,749 Truenat MTB tests were conducted and the unsuccessful tests were 1350. The invalid rate was 2.4%. A total of 119,247 Truenat assays were conducted during the study period with 2398 tests being unsuccessful giving an invalid rate of 2.0%.

Table 3 Total tests conducted and unsuccessful test results recorded (January to December 2022)
Full size table

Table 4 shows the prevalence of test errors. Less than a quarter of the health facilities, 23.7% had high DNA test errors. Also, less than one third of the health facilities, 31.6% had high MTB test errors.

Table 4 Prevalence of truenat high test errors
Full size table

Table 5 shows the factors associated with High errors with DNA extraction tests. Comparable proportions of respondents who were less than 38 years, 28.6% and those who were 38 years and above, 17.6% recorded high DNA test errors, (Fisher’s exact test, p = 0.476). A higher proportion of health facilities that are DOTS centers, 25.8% had high DNA extraction errors when compared with those that are not DOTS centers, 14.3% but the difference in proportions was not found to be statistically significant, (χ2 = 0.419, p = 0.517). A higher proportion of the health facilities that did not have the SOP for conducting Truenat pasted on the wall, 33.3% had high DNA extraction errors when compared with those who had the SOP pasted on the wall, 22.9%, but the difference in proportions was not found to be statistically significant, (Fisher’s exact test, p = 1.0). Comparable proportions of health facilities that had preventive maintenance log form visibly kept and recorded, 19.4% and those that did not, 42.9% had high DNA extraction errors, (χ2 = 1.745, p = 0.186).

Table 5 Factors associated with high errors with DNA extraction test
Full size table

Table 6 shows the factors associated with High errors with the MTB test. Comparable proportions of male, 29.6% and female respondents, 36.4% recorded high MTB test errors, (χ2 = 0.164, p = 0.658). Similarly, comparable proportions of health facilities that are microscopy centers, 31.6% and those that are not, 31.6% had high MTB test errors, (Fisher’s exact test, p = 1.0). A higher proportion of Laboratory staffs who were not trained by Molbio, 50.0% had high MTB test errors when compared with those who were trained by Molbio, 29.4% but the difference in proportions was not found to be statistically significant, (Fisher’s exact test, p = 0.577). A higher proportion of health facilities who did not have SOP for sample preparation for Truenat DNA extraction visibly pasted, 60.0% had high MTB test errors when compared with those who had the SOP pasted, 28.1% but the difference in proportions was not found to be statistically significant, (Fisher’s exact test, p = 0.357).

Table 6 Factors associated with high errors with MTB test
Full size table

Table 7 shows the distribution of unsuccessful Trueprep test results. A total of 1021 unsuccessful Trueprep test results were reported for all the health facilities. The highest error recorded was cartridge error, 450 while the least recorded error was device heater not working.

Table 7 Distribution of unsuccessful Trueprep test results from January to December 2022
Full size table

Table 8 shows the distribution of unsuccessful Truelab test results. A total of 1,198 unsuccessful Truelab test results were recorded for all the health facilities within the study period. The highest error recorded was internal control not amplifying in PCR, (933; 78%) while the least was the non-initialization of the Truenat machine (7; 0.6%).

Table 8 Distribution of unsuccessful Truelab test results from January to December, 2022
Full size table

Discussion

The Truenat implementation guidelines recommend that the target proportion of unsuccessful tests for Laboratory evaluation for TB should not exceed 3%5. The results from this study show low error rates of 1.8%, 2.4% and 0.1% from 56,799 Trueprep DNA extractions, 56,749 Truenat MTB tests and 5,699 Truenat MTB Rif tests respectively. This gives an average error rate of 2% for Truenat implementation in Nigeria and was found to be significantly lower when compared to results from other countries implementing Truenat. A study in Addis Ababa, Ethiopia, reported 6% unsuccessful test rates for Trueprep extractions, 11.1% non-determinate rate for Truenat MTB and 10% invalid rate for Truenat MTB Rif Dx assay8, while another multicenter study of four countries reported non-determinate rates of 2.4%, 6.2% and 2.6% for Trueprep, Truenat MTB, and Truenat MTB-RIF respectively9. In addition, the error rates from the Nigerian Truenat study were lower when compared to the 4.2% error rate reported from a study during the early roll-out of Gene Xpert in the country10. The low error rates reported in Nigeria may be related to the fact that > 80% of the Truenat sites had good adherence to the SOPs for Truenat implementation especially for preventive maintenance. It could also be inferred that the low error rates may be due to good quality of training and supportive supervision provided to the Laboratory staff.

In addition, the study showed that less than a quarter of the 38 Truenat sites in Nigeria, 23.7%, had high Trueprep DNA extraction error rates while less than a third of the facilities, 31.6%, had high Truenat MTB test errors within the first year of implementation. This could be considered a good outcome in the context of a number of factors which include; Truenat being a new tool introduced into the country’s diagnostic network, the more hands-on steps required for Truenat use compared to the GeneXpert and the fact that slightly above half of the laboratory end users in the study have had previous experience with any molecular diagnostic method. Although the study results showed that facilities with low error rates had better adherence to SOP for Truenat implementation especially for preventive maintenance, none of these was found to be significantly associated with the error rates. This may be due to the small sample size of the sites used for analysis.

The distribution of types of unsuccessful tests in the study showed that cartridge error (E2) and cartridge clogged (E3) accounted for about two-thirds of the unsuccessful Trueprep tests reported, with 44% of all unsuccessful Trueprep tests due to cartridge error (E2). The most frequent Truenat MTB error reported was invalid, (Internal control did not amplify in the PCR) which accounted for 78%. A review of the Molbio guidelines indicated that Trueprep cartridge errors and clogged are due to Incomplete liquefaction of the sputum samples during processing and infrequent flushing protocols, while the Truelab invalid errors causing Internal control not to amplify in the PCR is commonly due to Insufficient elute and dirty slider glass. Although there are no studies published in journals on the frequency of Truenat errors, the report from a Stop TB Truenat experience sharing session show that cartridge clogged (E3) was one of the common errors reported by the DRC during Truenat implementation as part of the new tool project11. In addition, a study on causes of errors in GeneXpert implementation in Ethiopia shows that Probe check errors due to sample processing issues was the most frequently reported error12. This therefore highlights areas of focus for additional training and supportive supervision.

Study limitation

All Trueprep and Truelab errors not recorded in the Truenat laboratory registers may not have been included in the analysis and may have affected the results of this study, especially the distribution of the errors. All Laboratory staff were encouraged to record all errors in the Truenat laboratory registers and these registers were reviewed on a weekly basis. In addition, the country is in the process of establishing diagnostics connectivity to all Truenat machines which will allow for real time monitoring and accurate records of error rates of Truelab instruments.

Conclusion

The implementation of Truenat as a new tool for TB diagnosis in Nigeria is generally associated with low error rates. Cartridge error and Internal control not amplifying in PCR were the commonest errors associated with Trueprep DNA extraction and Truelab MTB tests respectively. It is recommended that the Truenat training curriculum in-country be updated to focus on and address the possible underlying causes of these errors as the country plans for scale up. Continuous monitoring and supervision are also essential in identifying sites experiencing errors early for prompt intervention.