BACKGROUND AND OBJECTIVE: Pleural transudates are most commonly due to heart failure (HF) or hepatic hydrothorax (HH), but a number of these effusions are misclassified as exudates by standard (Light's) criteria. The aim of this study was to determine the prevalence of mislabelled transudates and to establish simple alternative parameters to correctly identify them. METHODS: We retrospectively analysed the pleural fluid and serum protein, lactate dehydrogenase and albumin concentrations from 364 cardiac effusions and 102 HH. The serum-to-pleural fluid protein and albumin gradients (serum concentration minus pleural fluid concentration), as well as the pleural fluid-to-serum albumin ratio (pleural fluid concentration divided by the serum concentration) were calculated for the mislabelled transudates. RESULTS: Light's criteria had misclassified more HF-associated effusions than HH (29% vs 18%, P = 0.002). A serum-to-pleural fluid protein gradient >3.1 g/dL correctly identified 55% and 61% of the HF and HH false exudates, respectively. The figures for an albumin gradient >1.2 g/dL were 83% and 62%. Finally, a pleural fluid-to-serum albumin ratio <0.6 had identical accuracy for labelling miscategorized cardiac and liver-related effusions (78% and 77%, respectively). CONCLUSIONS: If the clinical picture is consistent with HF but the pleural fluid meets Light's exudative criteria, the measurement of the albumin rather than the protein gradient is recommended. In the context of cirrhosis, a potentially 'false' exudate is identified better by the pleural fluid-to-serum albumin ratio.
BACKGROUND AND OBJECTIVE: Pleural transudates are most commonly due to heart failure (HF) or hepatic hydrothorax (HH), but a number of these effusions are misclassified as exudates by standard (Light's) criteria. The aim of this study was to determine the prevalence of mislabelled transudates and to establish simple alternative parameters to correctly identify them. METHODS: We retrospectively analysed the pleural fluid and serum protein, lactate dehydrogenase and albumin concentrations from 364 cardiac effusions and 102 HH. The serum-to-pleural fluid protein and albumin gradients (serum concentration minus pleural fluid concentration), as well as the pleural fluid-to-serum albumin ratio (pleural fluid concentration divided by the serum concentration) were calculated for the mislabelled transudates. RESULTS: Light's criteria had misclassified more HF-associated effusions than HH (29% vs 18%, P = 0.002). A serum-to-pleural fluid protein gradient >3.1 g/dL correctly identified 55% and 61% of the HF and HH false exudates, respectively. The figures for an albumin gradient >1.2 g/dL were 83% and 62%. Finally, a pleural fluid-to-serum albumin ratio <0.6 had identical accuracy for labelling miscategorized cardiac and liver-related effusions (78% and 77%, respectively). CONCLUSIONS: If the clinical picture is consistent with HF but the pleural fluid meets Light's exudative criteria, the measurement of the albumin rather than the protein gradient is recommended. In the context of cirrhosis, a potentially 'false' exudate is identified better by the pleural fluid-to-serum albumin ratio.