Abstract:
Tongue swabs have shown promise as alternatives to sputum for detecting Mycobacterium tuberculosis (MTB) in patients with pulmonary tuberculosis (TB). Some of the most promising results have come from studies that used manual quantitative PCR (qPCR) to analyze swabs. Studies using the automated Cepheid Xpert® MTB/RIF Ultra qPCR test (Xpert Ultra) to analyze tongue swabs have yielded more modest results. Using tongue swabs collected from TB-negative volunteers and spiked with cultured MTB cells, this work screened nearly 30 swabbing, storage, and extraction methods for using Xpert Ultra with tongue swabs. Three methods were then subjected to more rigorous evaluation. The most promising method, in which two COPAN FLOQswabs were collected and combined into a single tube and processed with Cepheid’s Sample Reagent, yielded a 95% limit of detection (LoD) of 76.5 CFU/swab (95% CI 53.7 – 103.4), with an error rate of 4.6%. This method exhibited suitable sensitivity and low error rate, and is prioritized for clinical evaluations. In comparison, the clinically validated manual qPCR method returned a LoD of 53.5 CFU/swab (95% CI 36.9 – 73.0), whereas Cepheid’s experiments in sputum Xpert Ultra yielded a LoD of 15.6 CFU/mL sputum (95% CI 12.2 – 23.1).