BACKGROUND: Diagnosing tuberculous pleural effusion (pTB) is often difficult because the culturing of tubercle bacilli results in a negative test in the majority of cases. Serological tests for the detection of antibodies to tuberculous glycolipid (TBGL) and lipoarabinomannan (LAM) have been introduced for the diagnosis of pulmonary tuberculosis. We examined the levels of these antibodies, adenosine deaminase (ADA) and interferon-gamma (IFN-gamma) in the pleural effusion and compared their diagnostic values in pTB. METHODS: We studied 65 patients with pleural effusion. Of those, 19 patients were diagnosed as having pTB according to our broad case definition. The etiologies in the other 46 patients were malignant effusion, transdative effusion and miscellaneous diseases. Determiner TBGL antibody (D-TBGL-Ab) and MycoDot were used for the detection of anti-LAM and anti-TBGL antibodies, respectively, in the pleural effusion. RESULTS: The sensitivity of ADA was 78.9% (15/19) and the specificity 97.8% (45/46). The sensitivity of IFN-gamma was 84.2% (16/19) and the specificity 93.5% (43/46). The sensitivities of D-TBGL-Ab and MycoDot were both 52.6% (10/19) and their specificities were 95.7% (44/46) and 97.8% (45/46), respectively. When DTBGL-Ab (cutoff point: 2.0 U/ml) and ADA activity (cutoff point: 57 IU/l) were combined, the sensitivity was 94.7% (18/19) and the specificity 93.5% (43/46). CONCLUSIONS: In the diagnosis of pTB, D-TBGL-Ab and MycoDot each have low sensitivity but high specificity. When D-TBGL-Ab is used in combination with ADA, the sensitivity and specificity are both >90%. We conclude that D-TBGL-Ab and ADA in combination are useful in the diagnosis of pTB.
BACKGROUND: Diagnosing tuberculous pleural effusion (pTB) is often difficult because the culturing of tubercle bacilli results in a negative test in the majority of cases. Serological tests for the detection of antibodies to tuberculous glycolipid (TBGL) and lipoarabinomannan (LAM) have been introduced for the diagnosis of pulmonary tuberculosis. We examined the levels of these antibodies, adenosine deaminase (ADA) and interferon-gamma (IFN-gamma) in the pleural effusion and compared their diagnostic values in pTB. METHODS: We studied 65 patients with pleural effusion. Of those, 19 patients were diagnosed as having pTB according to our broad case definition. The etiologies in the other 46 patients were malignant effusion, transdative effusion and miscellaneous diseases. Determiner TBGL antibody (D-TBGL-Ab) and MycoDot were used for the detection of anti-LAM and anti-TBGL antibodies, respectively, in the pleural effusion. RESULTS: The sensitivity of ADA was 78.9% (15/19) and the specificity 97.8% (45/46). The sensitivity of IFN-gamma was 84.2% (16/19) and the specificity 93.5% (43/46). The sensitivities of D-TBGL-Ab and MycoDot were both 52.6% (10/19) and their specificities were 95.7% (44/46) and 97.8% (45/46), respectively. When DTBGL-Ab (cutoff point: 2.0 U/ml) and ADA activity (cutoff point: 57 IU/l) were combined, the sensitivity was 94.7% (18/19) and the specificity 93.5% (43/46). CONCLUSIONS: In the diagnosis of pTB, D-TBGL-Ab and MycoDot each have low sensitivity but high specificity. When D-TBGL-Ab is used in combination with ADA, the sensitivity and specificity are both >90%. We conclude that D-TBGL-Ab and ADA in combination are useful in the diagnosis of pTB.