A prospective multicentre diagnostic accuracy study for the Truenat tuberculosis assays

Abstract

Background Bringing reliable and accurate tuberculosis (TB) diagnosis closer to patients is a key priority for global TB control. Molbio Diagnostics have developed the Truenat point-of-care molecular assays for detection of TB and rifampicin (RIF) resistance. Methods We conducted a prospective multicentre diagnostic accuracy study at 19 primary healthcare centres and seven reference laboratories in Peru, India, Ethiopia and Papua New Guinea to estimate the diagnostic accuracy of the point-of-care Truenat MTB, MTB Plus and MTB-RIF Dx assays for pulmonary TB using culture and phenotypic drug susceptibility testing as the reference standard, compared with Xpert MTB/RIF or Ultra. Results Of 1807 enrolled participants with TB signs/symptoms, 24% were culture-positive for Mycobacterium tuberculosis, of which 15% were RIF-resistant. In microscopy centres, the pooled sensitivity of Truenat MTB and Truenat MTB Plus was 73% (95% CI 67–78%) and 80% (95% CI 75–84%), respectively. Among smear-negative specimens, sensitivities were 36% (95% CI 27–47%) and 47% (95% CI 37–58%), respectively. Sensitivity of Truenat MTB-RIF was 84% (95% CI 62–95%). Truenat assays showed high specificity. Head-to-head comparison in the central reference laboratories suggested that the Truenat assays have similar performance to Xpert MTB/RIF. Conclusion We found the performance of Molbio’s Truenat MTB, MTB Plus and MTB-RIF Dx assays to be comparable to that of the Xpert MTB/RIF assay. Performing the Truenat tests in primary healthcare centres with very limited infrastructure was feasible. These data supported the development of a World Health Organization policy recommendation of the Molbio assays.