Xpert MTB/RIF assay as an initial diagnostic test in pulmonary tuberculosis: a multicenter prospective randomized pragmatic trial

Nucleic acid amplification tests (NAATs) were performed selectively as an initial diagnostic test for pulmonary tuberculosis (TB) in Taiwan. We assessed whether expanded use of Xpert MTB/RIF assay as an initial diagnostic test made positive contribution to the management of TB. Patients who had NAATs as an initial diagnostic test requested by clinician were classified as group A. Those who did not were randomized in a 1:1 ratio into group B who had an immediate Xpert test by intervention and group C who had usual care. 6835 patients were enrolled. Comparing group B and group C, the proportion of patients diagnosed with active TB (3.1% vs 2.7%, p = 0.336), the proportion of patients died before anti-TB treatment (2.3% vs 5.1%. p = 0.318), the median interval between submitting sputum and initiation of anti-TB treatment (7.0 days, interquartile range 3.0-25.0 vs 6.0 days, interquartile range 2.0-23.0, p = 0.589), and the proportion of TB patients with treatment success (73.8% vs 81.8%, p = 0.657) were not significantly different. There is no evidence that expanded use of Xpert MTB/RIF test as an initial diagnostic test for pulmonary TB among cases with a relatively low pretest probability of TB has positive influence on TB control in Taiwan.Trial Registration: ClinicalTrials.gov. Number NCT04433195 (date: 16/06/2020).