Diagnosing tuberculous pleural effusion using clinical data and pleural fluid analysis A study of patients less than 40 years-old in an area with a high incidence of tuberculosis.

BACKGROUND: Tuberculous pleural effusions (TPE) are common. The diagnosis is often problematic. As the determination of ADA is often unavailable in some countries, the aim of this study was to evaluate the diagnostic usefulness of other data from pleural fluid analysis, in young patients from populations with high prevalence of tuberculosis (TB).
METHODS: We analysed 218 patients with pleural effusion (165 tuberculous, 21 infectious, 11 neoplastic, 16 miscellaneous, 3 idiopathic). We performed two regression models; one included pleural fluid ADA values (model 1), and the other without ADA (model 2).
RESULTS: Model 1 selected two variables (ADA >35U/L) and lymphocytes (>31.5%) and correctly classified 216/218 effusions (1 false negative, 1 false positive). Model 2 (without ADA) selected three variables: lymphocytes (>31.5%), fever and cough, and correctly classified 207/218 effusions (8 false negatives, 3 false positives). The sensitivity of models 1 and 2 was 99.4% and 95.2%, specificity 98.1% and 94.3% and accuracy 99% and 95%.
CONCLUSIONS: In geographic areas with high prevalence of TB and a low prevalence of HIV, in young patients (<or=40 years), it is possible to confidently diagnose TPE with either of the two regression tree models, with the utility of ADA providing superior sensitivity, specificity, and accuracy. Copyright 2010 Elsevier Ltd. All rights reserved.