Diagnostic value of gastric aspirate smear and polymerase chain reaction in smear-negative pulmonary tuberculosis.

Abstract

The aim of this study was to determine the validity of acid-fast bacilli (AFB) smear and polymerase chain reaction (PCR) from gastric aspirates for the diagnosis of smear-negative pulmonary tuberculosis. A cross-sectional study was conducted in a university hospital. One hundred and nine patients with suspected pulmonary tuberculosis in whom either sputum smears were negative or who were not producing sputum were recruited to the study. All patients underwent gastric aspiration after an overnight fast followed by standard fibreoptic bronchoscopy. Specimens were subjected to AFB smear, culture, and pathological examination. PCR was performed on culture filtrate after 1 week of incubation. Eight patients did not complete the follow-up schedule. Of the 101 patients with final outcomes, a diagnosis of pulmonary tuberculosis from microbiological evidence was established in 54 patients. The gastric aspirate smear, PCR, or either one of them was positive in 34, 30, and 39 tuberculosis patients, respectively. There were 13 false positive smears from 47 non-tuberculosis patients, with five resulting from non-tuberculous mycobacteria (NTM). The PCR was falsely positive in eight patients, five of whom had previous histories of tuberculosis. The overall sensitivity, specificity, positive predictive value, and negative predictive value of gastric aspirate examination by combined smear and PCR were 72, 58, 66, and 64%, respectively. Gastric aspiration is a useful tool for the diagnosis of smear-negative pulmonary tuberculosis warranting institution of antituberculosis treatment. Interpretation of the results should be cautious in those who have had tuberculosis in the past or who have been at risk for acquisition of NTM.