Pulmonary tuberculosis (TB) is an important cause of morbidity and mortality especially in developing countries. A definitive microbiologic confirmation of Mycobacterium tuberculosis is important in the diagnosis of childhood TB. We aimed to compare the diagnostic value of nasopharyngeal aspirate (NPA), gastric aspirate (GA) and bronchoalveolar lavage (BAL) specimens in children with highly suspected pulmonary tuberculosis (TB). NPA, GA and BAL samples were obtained from forty patients. The mean age was 9.2±4.7 years. Sixty-eight percent of children had a history of household contact and 82% had tuberculin skin test positivity. Acid-fast bacilli (AFB) stain was positive in 22.5% (N=9) of BAL, 17.5% (N=7) of GA, and 10% (N=4) of NPA samples. Positive Lowenstein-Jensen culture was 27.5% (N=11) in BAL, 22.5% (N=9) in GA, and 12.5% (N=5) in NPA samples. Positive AFB stains and growth in TB cultures from BAL fluid and GA samples were both higher than NPA samples (p < 0.006 and p < 0.004, respectively GA). We conclude that NPA specimen fails to determine Mycobacterium tuberculosis in children with highly suspected pulmonary TB when compared to GA or BAL fluid.

Keywords: Mycobacterium tuberculosis, bronchoalveolar lavage fluid, childhood pulmonary tuberculosis, diagnosis, gastric aspirate, nasopharyngeal aspirate