Search

Tuberculosis (TB) remains one of the leading causes of death worldwide and is caused by the single infectious agent Mycobacterium tuberculosis (Mtb). Although sputum is the most common specimen for pulmonary TB detection, some other respiratory specimens, such as bronchoalveolar lavage (BAL) fluid, gastric lavage (GL), and induced sputum (IS), are also collected from patients who are unable to deliver sputum. In this study, we aimed to evaluate the diagnostic performances of different test methods for TB diagnosis using BAL fluid specimens from sputum-scarce pulmonary TB patients. In this current study, a total of 210 BAL fluid specimens were collected and subjected to culture on Lowenstein–Jensen (L-J) medium, using an N-acetyl-L-cysteine-Sodium Hydroxide decontamination and digestion method, Xpert MTB/RIF (Xpert, Cepheid, Sunnyvale, CA, USA) assay, and acid-fast bacilli (AFB) microscopy with a Ziehl–Neelsen staining method for the detection of pulmonary TB. The sensitivity and specificity of these methods were then analyzed against the composite reference standard (CRS). Additionally, the receiver operating characteristic (ROC) curve was used to evaluate the diagnostic value of these assays. Among the 210 specimens, 39 (18.6%), 27 (12.8%), and 12 (5.7%) were found positive with Xpert assay, culture, and AFB microscopy, respectively. Considering the CRS, 42 (20%) were positive as the final diagnosis. The Xpert assay had a significantly higher sensitivity (92.9%, 95% CI: 80.5–98.5) compared to culture (64.3%, 95% CI: 48.0–78.4) and AFB microscopy (28.6%, 95% CI: 15.7–44.6) against the CRS. Additionally, the area under the ROC curve (AUC) for the Xpert assay, culture, and AFB microscopy accounted for 0.964, 0.821, and 0.655, respectively, when using CRS as the reference. In conclusion, our study findings demonstrated that the Xpert assay conferred a considerable diagnostic potential compared to other conventional methods for the diagnosis of pulmonary TB from BAL fluid specimens. [ABSTRACT FROM AUTHOR]