RESUMO
BACKGROUND AND OBJECTIVE: Light's criteria are frequently used to evaluate the exudative or transudative nature of pleural effusions. However, misclassification resulting from the use of Light's criteria has been reported, especially in the setting of diuretic use in patients with heart failure (HF). The objective of this study was to evaluate the utility of B-type natriuretic peptide (BNP) measurements as a diagnostic tool for determining the cardiac aetiology of pleural effusions. METHODS: Patients with pleural effusions attributable to HF (n = 34), hepatic hydrothorax (n = 10), pleural effusions due to cancer (n = 21) and pleural effusions due to tuberculosis (n = 12) were studied. Diagnostic thoracentesis was performed for all 77 patients. Receiver operating characteristic (ROC) curves were constructed to determine the diagnostic accuracy of plasma BNP and pleural fluid BNP for the prediction of HF. RESULTS: The areas under the ROC curves were 0.987 (95% CI 0.93-0.998) for plasma BNP and 0.949 (95% CI 0.874-0.986) for pleural fluid BNP, for distinguishing between patients with pleural effusions caused by HF (n = 34) and those with pleural effusions attributable to other causes (n = 43). The cut-off concentrations with the highest diagnostic accuracy for the diagnosis of HF as the cause of pleural effusion were 132 pg/mL for plasma BNP (sensitivity 97.1%, specificity 97.4%) and 127 pg/mL for pleural fluid BNP (sensitivity 97.1%, specificity 87.8%). CONCLUSIONS: In patients with pleural effusions of suspected cardiac origin, measurements of BNP in plasma and pleural fluid may be useful for the diagnosis of HF as the underlying cause.