Comparison of Xpert MTB RIF (G4) With Xpert MTB (Ultra) in Diagnosis of Mycobacterium Tuberculosis in Clinical Samples

Abstract Background: Tuberculosis (TB) continues to be a major cause of morbidity and mortality worldwide. The main strategy to fight the pandemic is the development of rapid, accurate diagnostic tests, which helps to reduce the time for initiation of therapy. In 2010, WHO recommended the use of Xpert Mycobacterium tuberculosis /rifampicin (MTB/RIF) (G4), which is a cartridge-based semi-automated nucleic acid amplification test (NAAT) as a primary test with an aim to improve the detection of MTB. Later on, to improve the diagnostic efficacy, the new and improved cartridges called Xpert Ultra were introduced in 2017. Materials and Methods: One thousand and sixty-nine clinical samples were utilized for the study for microscopy, culture, and NAATs (Xpert MTB/RIF G4 and Ultra). The samples were randomly allocated. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated while assuming trace results as positive and negative in two separate scenarios. Results: The sensitivity of Xpert MTB/RIF Ultra cartridge was higher (100%) than Xpert MTB/RIF G4 cartridge (95.23%) and the specificity of Ultra cartridge was lower (71.79%) when compared to G4 cartridge (91.63%). However, the specificity of Xpert Ultra cartridge improved by reclassifying the “trace” interpretation as negative (89.1%). The NPV was 100% and 99.09% for Xpert Ultra cartridge and the Xpert G4 cartridge, respectively, and did not change by reclassifying the “trace” interpretation as negative. In contrast, the PPV varied greatly between the two tests, 26.66% for Xpert Ultra cartridge and 66.66% for Xpert G4 cartridge. The PPV of Xpert Ultra cartridge was improved to 48.48% by including “trace” samples in the negative pool. Conclusion: Xpert Ultra and G4 shows good performance as a first line diagnostic test. Caution should be exercised when interpreting ‘trace’ results in a country like India without parallel confirmatory tests like microscopy and culture. Data is limited on this newly incorporated value assigned in the automated report. Conducting further studies on results interpreted as ‘trace’ will help in improving the guidelines and as well as utilisation of such tests.