Abstract
The diagnosis of pleural tuberculosis (TB) is problematic. The comparative performance of newer same-day tools for pleural TB, including Xpert MTB/RIF Ultra (ULTRA), has hitherto not been comprehensively studied. Adenosine deaminase (ADA), IRISA-TB (interferon gamma ultrasensitive rapid immunosuspension assay), Xpert MTB/RIF, and ULTRA performance outcomes were evaluated in pleural fluid samples from 149 patients with suspected pleural TB. The reference standard was culture positivity (fluid, biopsy specimen, or sputum) and/or pleural biopsy histopathology (termed definite TB). Those designated as having non-TB were negative by microbiological testing and were not initiated on anti-TB treatment. To determine the effect of sample concentration, 65 samples underwent pelleting by centrifugation, followed by conventional Xpert MTB/RIF and ULTRA. Of the 149 patients, 49 had definite TB, 16 had probable TB (not definite but treated for TB), and 84 had non-TB. ULTRA sensitivity and specificity (95% confidence intervals [CI]) were similar to those of Xpert MTB/RIF [sensitivity, 37.5% (25.3 to 51.2) versus 28.6% (15.9 to 41.2), respectively; specificity, 98.8% (96.5 to 100) versus 98.8% (96.5 to 100), respectively]. Centrifugation did not significantly improve ULTRA sensitivity (29.5% versus 31.3%, respectively). Adenosine deaminase and IRISA-TB sensitivity were 84.4% (73.9 to 95.0) and 89.8% (81.3 to 98.3), respectively. However, IRISA-TB demonstrated significantly better specificity (96.4% versus 87.5% [P = 0.034]), positive predictive value (93.6% versus 80.9 [P = 0.028]), and positive likelihood ratio (25.1 versus 6.8 [P = 0.032]) than ADA. In summary, Xpert ULTRA has poor sensitivity for the diagnosis of pleural TB. Alternative assays (ADA and IRISA-TB) are significantly more sensitive, with IRISA-TB demonstrating a higher specificity and rule-in value than ADA in this high-TB-burden setting where HIV is endemic.