Differential diagnosis of tuberculous and malignant pleural effusions: what is the role of adenosine deaminase?

Abstract

The objective of this study was to evaluate the utility of invasive and noninvasive diagnostic procedures in tuberculous pleurisy (TPE) in an area with intermediate incidence of tuberculosis. The aim was to determine the cutoff value for adenosine deaminase (ADA) and the sensitivity and specificity of ADA and evaluate pleural fluid cytology and pleural biopsy in the differential diagnosis of malignant and tuberculous pleurisy. The study included 121 patients. TPE was confirmed in 54 patients and malignant effusion in 67 patients. Criteria used for TPE diagnosis were positive cultures of effusion or biopsy specimen, tuberculous granulomas, or positive sputum cultures without other explanation for pleural effusion. Malignancy was diagnosed by either cytology or biopsy. The cutoff value of ADA in TPE was 49 U/L, sensitivity was 89.2%, specificity was 70.4%, positive predictive value (PPV) was 84.4%, and negative predictive value (NPV) was 78.4%. ADA activity below 16 U/L suggests that TPE is highly unlikely with sensitivity=38.5%, specificity=100%, PPV=100%, and NPV=57.4%. ADA effusion/serum ratio reached a cutoff in TPE of 1.7 (sensitivity=84.6%, specificity=72.2%, PPV=81.4%, NPV=71.4%). Sensitivity, specificity, PPV, and NPV of cytology evaluation for TPE are 72.2%, 70.1%, 66.1%, and 75.8%, respectively. Pleuroscopy-guided pleural biopsy had sensitivity=66.7%, specificity=100%, PPV=100%, and NPV=78.8%. In 27.8% of TPE cases, pleural fluid cultures were positive. There is no doubt that pleuroscopy-guided biopsy is of great value for TPE diagnosis; however, sensitivity and specificity of noninvasive tests, especially ADA, can help to distinguish between TB and malignancy.