The Diagnostic Usefulness of Pleural Fluid Adenosine deaminase with Lymphocyte/Neutrophil Ratio in Tuberculous Pleural Effusion

2004
Background : The measurement of adenosine deaminase( ADA) level in pleural fluid is useful in the diagnosis of tuberculous(TB) pleural effusion. However, ADAis also elevated in other diseases such as malignancy, bacterial infections, empyema, and collagen vascular disease, ADAalone has limited value. The object of this study is to determine diagnostic usefulness of the combined use of ADAvalue with lymphocyte/neutrophil ratio(L/N ratio) rather than the use of ADAalone. Method : We evaluated 198 patients(age=55.9±12.9, M/F=2.7:1) with pleural effusionwho had admitted in Gyeong? sang National University Hospital from Jan. 1999 to Dec. 2001. retrospectively. Patients were divided into four diagnostic groups: TB pleural effusion(n=91), parapneumonic effusion(n=65), malignant effusion(n=21), and transudative effusion(n=13). The ADAlevel, differential cell count, biochemistry, cytology, and microbiology of each diagnostic groups were evaluated. The sensitivity, specificity, negative predictive value(npv), positive predictive value(ppv) and efficiency were calculated at each ADAvalues and combined ADAvalue with various L/N ratios. Results : The ADAlevel in TB pleural effusionwas significantly higher than that of parapneumonic effusion, malignant pleural effusion, and transudative effusion(p<0.05). Sensitivity, specificity, ppv, npv and efficiency at ADA≧50 IU/L in the diagnosis of TB pleural effusionwere 89.0%, 82.2%, 81.0%, 89.8% and 85.5% respectively. When ADA≧50 IU/L was combined with lymphocyte/neutrophil ratio≧0.75, sensitivity, specificity, ppv, npv, and efficiency were 83.5%, 96.3%, 95.0%, 87.9% and 90.5% respectively. Specificity, ppv and efficiency were increased with combination of ADAvalue and L/N ratio. Conclusion : Combination of ADAvalue and L/N ratio in pleural effusionis more useful than ADAvalue alone in the diagnosis of TB pleural effusion. (Tuberc Respir Dis 2004; 57:132-137)
    • Correction
    • Source
    • Cite
    • Save
    16
    References
    5
    Citations
    NaN
    KQI
    []
    Baidu
    map