Clinical performance of Xpert Mycobacterium tuberculosis/rifampin assay for diagnosis of pulmonary tuberculosis in patients with acquired immunodeficiency syndrome / 中华传染病杂志

ABSTRACT

Objective:To investigate the diagnostic performance of Xpert Mycobacterium tuberculosis/rifampin (Xpert MTB/RIF) assay for pulmonary tuberculosis (TB) in patients with acquired immunodeficiency syndrome (AIDS). Methods:Clinical data of 226 patients with AIDS and suspected pulmonary TB in Shanghai Public Health Clinical Center, Fudan University from July 2017 to November 2019 were retrospectively analyzed. Fluorescence staining microscopy of sputum smear, BACTEC MGIT 960 liquid culture (or Roche solid culture) and Xpert MTB/RIF assay were implemented respectively. The sensitivity and specificity of Xpert MTB/RIF in the diagnosis of Mycobacterium tuberculosis (MTB) infection and rifampin resistance were analyzed. Results:Totally 226 patients of suspected pulmonary TB were enrolled. There were 94(41.6%) patients had positive mycobacterium culture, in which 51 (54.3%) were MTB and 43 (45.7%) were nontuberculous mycobacteria (NTM). Using the positive MTB culture of sputum and mycobacterial protein from BCG of Rm 0.64 in electrophoresis (MPB64) as reference standard, the sensitivity and specificity of Xpert MTB/RIF assay for MTB diagnosis were 72.6%(95% confidence interval ( CI) 66.7%-78.4%) and 97.1% (95% CI 95.0%-99.3%), respectively. The sensitivity and specificity of Xpert MTB/RIF assay for MTB diagnosis in patients with positive sputum smear were 76.7%(95% CI 67.7%-85.8%) and 90.0(95% CI 83.6%-96.5%), respectively. The sensitivity and specificity of Xpert MTB/RIF assay for MTB diagnosis in patients with negative sputum smear were 50.0%(95% CI 41.8%-58.2%)and 99.3%(95% CI 97.9%-100.0%), respectively. With phenotypic resistance as reference standard, the sensitivity and specificity of Xpert MTB/RIF assay for rifampicin resistance were 75.0% and 100.0%, respectively. Conclusion:Among AIDS patients, the performance of Xpert MTB/RIF assay for pulmonary TB diagnosis is pretty good and could differentiate MTB from NTM rapidly, which has good application value.