Evaluation of the Xpert® MTB/RIF assay and its impact on tuberculosis diagnosis and rifampicin resistance screening: Efforts to determine the prevalence of drug Resistant TB and rpob gene mutations in the Malawian adult population

Abstract

Tuberculosis (TB) remains a global challenge and continues to kill more people than any other bacterial infection. The problem has been complicated with the emergence of multi-drug resistant (MDR) and extensively drug-resistant (XDR) TB. The Xpert MTB/RIF, which detects both Mycobacterium tuberculosis (MTB) and resistance to rifampicin (RIF) in a single rapid assay, has greatly improved TB diagnosis and detection of drug resistance. The main objective of the study was to determine the prevalence of resistance to first line TB drugs among outpatients initiating on treatment in Lilongwe, evaluate performance of Xpert MTB/RIF and to determine the prevalence of mutations responsible for RIF resistance in the rpoB gene of MTB strains from Malawi. This was achieved by analysing specimens collected from Bwaila Hospital (Bwaila) and the National TB Reference Laboratory (NTRL), both from Lilongwe district in Malawi. An observational cohort study was conducted between April 2011 and July 2012 at Bwaila to determine the prevalence of drug-resistant TB among outpatients. The study collected specimens from 702 TB patients of which 420 (59.8%) were HIV co-infected. One (1/349, 0.3%) RIF mono-resistant TB case was detected while isoniazid mono-resistance was detected in 2.9% (10/349) of the newly treated and culture confirmed cases. The prevalence of MDR-TB was 3.8% (1/26) among previously treated cases and 0.3% (1/349) among new cases. This demonstrates a very low prevalence of drug-resistant TB in the outpatient population. For TB diagnosis, the performance of Xpert MTB/RIF was evaluated retrospectively using 351 specimens (frozen stored pellets) from this cohort. Of 348 sputum pellets with valid Xpert MTB/RIF results, 200 (57%) were from HIV/TB co-infected while 148 (43%) were from HIV uninfected individuals. Among these, 219 (63%) were smear-negative and 129 (37%) were smear-positive. When compared to culture, Xpert MTB/RIF demonstrated a sensitivity of 93.8% (95% CI: 89.4% - 96.8%) and specificity of 97.4% (95% CI: 93.5% - 99.3%) with a positive predictive value of 97.8% (95% CI 94.6% - 99.4%) and negative predictive value of 92.6% (95% CI 87.4% - 96.1%). Within this cohort, Xpert MTB/RIF correctly identified 185 of 186 (98.8%) RIF-sensitive MTB and 2/2 (100%) RIF-resistant MTB when compared with MTBDRplus version 2. DNA sequencing resolved the one discrepant result in favour of MTBDRplus. In addition, Xpert correctly identified four non-tuberculous mycobacteria (NTM) [M. avium (n=1), M. intracellulare (n=3)] as negative for MTB. To determine the prevalence of mutations in the rpoB gene, the Rifampicin Resistance Determining Region (codon 507 – 533) was sequenced from specimens from both Bwaila and the National Tuberculosis Reference Laboratory (NTRL). Specimens that were reported with RIF resistance by Xpert MTB/RIF were eligible for DNA sequencing. A total of 43 specimens were successfully sequenced and mutations were detected in 37/43 specimens. Mutations were frequently detected in codons 526, 531 and 516. Mutations were not detected in 6/43 (14%) specimens. Among the specimens with no mutations, 2/6 (33%) were RIF-sensitive while 4/6 (67%) were RIF-resistant on MTBDRplus assay. Overall, the prevalence of MDR-TB is low among outpatients seeking treatment at Bwaila which emphasises the effectiveness of the National TB programme in Malawi. The use of Xpert MTB/RIF in this population can increase the number of patients starting on appropriate TB treatment earlier and offer a greater opportunity in increasing TB case detection among high risk and vulnerable groups. This technology can reduce the TB burden in the population including people living with HIV. Disease transmission can be reduced with the use of Xpert MTB/RIF because of early treatment initiation and therefore likely to shorten the sputum culture conversion times for patients with RIF-resistant TB. Xpert MTB/RIF is a useful tool due to its diagnostic accuracy although our findings show a lower sensitivity in smear-negative/culture-positive specimens. Further evaluations of Xpert MTB/RIF assay’s performance are required in detecting smear-negative/Xpert-positive cases to gauge its diagnostic value over a longer time period.