Comparison of diagnostic efficacy of three different oral samples in pulmonary tuberculosis using cepheid gene Xpert® MTB/RIF ultra.

CONTEXT: The etiological confirmation of pulmonary tuberculosis typically relies on sputum or bronchoalveolar lavage fluid (BALF). Oral sampling offers a non-invasive and non-sputum alternative, but evidence varies widely due to inconsistent methods. This study employed the Cepheid Gene Xpert® MTB/RIF Ultra (Xpert-Ultra) to compare the diagnostic performance of tongue swabs, pharyngeal swabs, and posterior oropharyngeal saliva (POS) collected from the same patient to identify the optimal option.

METHODS: The diagnosis was based on a composite microbiological reference standard. Tongue swabs, pharyngeal swabs, and POS were sequentially collected from each participant. Sputum and BALF underwent acid-fast bacilli smear microscopy, Cepheid Gene Xpert® MTB/RIF (Xpert), and culture, while oral samples were tested with Xpert-Ultra.

RESULTS: All three oral sample types demonstrated 100 % specificity. Compared to tongue swabs and pharyngeal swabs, POS showed the highest sensitivity, both overall and in subgroup analyses, along with the highest bacterial load and the lowest cycle threshold values.

CONCLUSION: Compared with tongue swabs and pharyngeal swabs, POS exhibits the best diagnostic efficacy, and is more suitable for no-sputum or paucibacillary tuberculosis patients, and shows potential as an alternative to BALF.