Usefulness of Mycobacterium tuberculosis-polymerase chain reaction with bronchial washing samples in predicting discontinuation of airborne infection isolation in patients hospitalized with suspected pulmonary tuberculosis.

ObjectiveIn-hospital tuberculosis (TB) transmission remains a concern. Airborne infection isolation (AII) can be discontinued in hospitalized patients with suspected active pulmonary TB when the results of three consecutive sputum acid-fast bacilli (AFB) smears are negative. However, fiberoptic bronchoscopy can be performed in patients who may have difficulty in producing sputum samples. This study aimed to investigate the usefulness of Mycobacterium tuberculosis-polymerase chain reaction (MTB-PCR) with bronchial washing specimens in predicting AII discontinuation in hospitalized patients with suspected active pulmonary TB.MethodsWe reviewed the medical charts of patients admitted to a tertiary hospital who were isolated and underwent fiberoptic bronchoscopy for suspicious pulmonary TB from January 2016 to December 2019. Patients with positive MTB-PCR results in the initial sputum examination were excluded. Criteria for discontinuing AII were defined as negative results for three consecutive AFB smears from respiratory specimens, or cases diagnosed other than TB. The study patients were divided into two groups: TB group and non-TB group.ResultsIn total, 166 patients were enrolled in the study. Of them, 35 patients were diagnosed with TB. There was no significant difference between the number of males in the TB (81; 61.8%) and non-TB (21; 60.0%) group. Though 139 patients had negative results on MTB-PCR using washing specimens, eight showed positive AFB culture. Of the 139 patients with negative MTB-PCR results, 138 had negative results for three consecutive AFB smears or were established to not have pulmonary TB. Therefore, the predictive accuracy of MTB-PCR with bronchial washing samples for discontinuing AII was 99.2%.ConclusionAlthough a negative result from MTB-PCR with bronchial washing samples cannot exclude pulmonary TB, it can predict AII discontinuation in hospitalized patients with suspected active pulmonary TB.