Background: Recently, much interest has been focused on the quantification of DNA in miscellaneous body fluids. In this study, the application is extended to classifying pleural effusions by measuring cell-free DNA in pleural fluid.
Methods: We recruited 50 consecutive patients with pleural effusions with informed consent. Pleural fluids were centrifuged at 13000 g, with supernatants aliquoted for extraction and analysis of beta-globin DNA sequence by quantitative real-time PCR. Serum and pleural fluid biochemistries were performed to classify pleural effusions using the modified criteria of Light et al. (Ann Intern Med 1972;77:507-13). The ROC curve was plotted to determine the cutoff DNA concentration for classifying pleural fluids as transudates or exudates. Indicators of diagnostic accuracy were calculated for both pleural fluid DNA and modified criteria of Light et al., using the discharge, microbiologic, and histologic diagnoses as the reference standard.
Results: The area under the ROC curve was 0.95 [95% confidence interval (CI), 0.84-0.99]. At 509 genome-equivalents/mL, pleural fluid DNA alone correctly classified 46 of 50 pleural effusions with 91% sensitivity (95% CI, 76-98%), 88% specificity (95% CI, 64-98%), and positive and negative likelihood ratios of 7.7 (95% CI, 3.1-19.5) and 0.10 (95% CI, 0.04-0.27), respectively. With the modified criteria of Light et al., 43 of 50 pleural effusions were correctly classified with 97% sensitivity (95% CI, 91-100%) and 67% specificity (95% CI, 45-89%). There were significant correlations between cell-free DNA and both lactate dehydrogenase and total protein in pleural fluid, suggesting their common origin.
Conclusions: Pleural fluid DNA concentrations are markedly increased in exudative effusions, making it a potential new tool to evaluate the etiologic causes of pleural effusions.