SETTING: University Hospital, Bangkok, Thailand. OBJECTIVE: To evaluate the diagnostic value of antibody detection in serum and in pleural effusion as a marker of tuberculous pleuritis (TBP). DESIGN: Cross-sectional study. MATERIALS AND METHODS: One hundred and fifty-five patients with pleural effusion who underwent diagnostic evaluation at Siriraj Hospital between March 1999 and May 2000 were recruited. Samples of pleural fluid were examined biochemically, cytologically and microbiologically. Pathological examination of pleural tissue was also performed. The diagnosis of TBP or other diagnosis was made by either pathological finding or culture result. Immunochromatographic tuberculosis (ICT-TB) tests for antibody detection were then performed using the stored serum samples and effusions from those patients with a final definite diagnosis. This test detects antibodies to five secreted antigens of Mycobacterium tuberculosis, including the 38 kDa antigen. RESULTS: We investigated 67 patients with TBP, 44 with malignant pleural effusions, seven with transudates and one with cryptococcal pleuritis. The combined ICT-TB serum and effusion tests were positive in 34/67 TBP and 22/52 non-TBP patients. The sensitivity, specificity, positive predictive value and negative predictive value of the ICT-TB test were 50.7, 57.7, 60.7 and 47.6%, respectively. In 11 TBP patients with human immunodeficiency virus (HIV) co-infection, the sensitivity of the ICT-TB test was 45.6%. There was no correlation between the test positivity and culture result or duration of disease. CONCLUSIONS: The diagnostic value of antibody detection in TBP is modest in an area with intermediate prevalence of tuberculosis, independently of HIV serological status.
SETTING: University Hospital, Bangkok, Thailand. OBJECTIVE: To evaluate the diagnostic value of antibody detection in serum and in pleural effusion as a marker of tuberculous pleuritis (TBP). DESIGN: Cross-sectional study. MATERIALS AND METHODS: One hundred and fifty-five patients with pleural effusion who underwent diagnostic evaluation at Siriraj Hospital between March 1999 and May 2000 were recruited. Samples of pleural fluid were examined biochemically, cytologically and microbiologically. Pathological examination of pleural tissue was also performed. The diagnosis of TBP or other diagnosis was made by either pathological finding or culture result. Immunochromatographic tuberculosis (ICT-TB) tests for antibody detection were then performed using the stored serum samples and effusions from those patients with a final definite diagnosis. This test detects antibodies to five secreted antigens of Mycobacterium tuberculosis, including the 38 kDa antigen. RESULTS: We investigated 67 patients with TBP, 44 with malignant pleural effusions, seven with transudates and one with cryptococcal pleuritis. The combined ICT-TB serum and effusion tests were positive in 34/67 TBP and 22/52 non-TBP patients. The sensitivity, specificity, positive predictive value and negative predictive value of the ICT-TB test were 50.7, 57.7, 60.7 and 47.6%, respectively. In 11 TBP patients with human immunodeficiency virus (HIV) co-infection, the sensitivity of the ICT-TB test was 45.6%. There was no correlation between the test positivity and culture result or duration of disease. CONCLUSIONS: The diagnostic value of antibody detection in TBP is modest in an area with intermediate prevalence of tuberculosis, independently of HIV serological status.
Authors: Karen R Steingart; Laura L Flores; Nandini Dendukuri; Ian Schiller; Suman Laal; Andrew Ramsay; Philip C Hopewell; Madhukar Pai Journal: PLoS Med Date: 2011-08-09 Impact factor: 11.069