Use of bronchoalveolar lavage enzyme-linked immunospot for diagnosis of smear-negative pulmonary tuberculosis

Abstract

To compare the diagnostic validity of blood enzyme-linked immunospot assay (ELISpot), broncho-alveolar lavage (BAL) ELISpot and the tuberculin skin test (TST) in patients with pulmonary smear-negative tuberculosis (TB) in a country with high TB prevalence. In a prospective, hospital-based study, 107 patients with suspected TB were tested simultaneously using blood and BAL ELISpot and TST. Of 102 patients with active pulmonary TB, 36 (35.3%) were diagnosed with TB, while 66/102 (64.7%) had a non-TB diagnosis. The sensitivity and specificity for ELISpot on mononuclear cells from BAL fluid was respectively 94.4% (95%CI 81.9-98.5) and 78.1% (95%CI 66.6-86.5). The specificity of BAL ELISpot was significantly higher than that of blood ELISpot (P = 0.011). Compared with blood ELISpot and TST, BAL ELISpot was not significantly influenced by previous history of TB (OR 2.05, P > 0.05) or household contact with a patient with active TB (OR 2.41, P > 0.05). ELISpot on BAL appears to be a more rapid and sensitive supplementary test than on blood for the diagnosis of active TB patients with a negative sputum smear in a developing country setting with high TB prevalence and access to bronchoscopy and ELISpot assay. However, the test's utility was limited by its moderate specificity.