BACKGROUND AND OBJECTIVE: The purpose of this study was to compare the diagnostic utility of pleural fluid N-terminal pro-B-type natriuretic peptide (NT-proBNP), midregion pro-atrial natriuretic peptide (MR-proANP) and midregion pro-adrenomedullin (MR-proADM) for discriminating heart failure (HF)-associated effusions. METHODS: NT-proBNP, MR-proANP and MR-proADM were measured by commercially available methodologies in the pleural fluid of a retrospective cohort of 185 consecutive patients with pleural effusions, of whom 95 had acute decompensated HF. Receiver-operating characteristic and area under the curve (AUC) analyses allowed comparisons of the discriminative properties of these biomarkers to be made at their optimal cut-off points. RESULTS: The diagnostic accuracy of NT-proBNP and MR-proANP for HF as quantified by the AUC was 0.935 and 0.918, respectively, whereas MR-proADM was of limited value (AUC = 0.62). A pleural fluid MR-proANP >260 pmol/L or NT-proBNP >1700 pg/mL argues for HF (likelihood ratio (LR) positive >5), while levels below these cut-off values significantly decrease the probability of having the disease (respective LR negative 0.19 and 0.10). The optimal cut-off points for natriuretic peptides were influenced by age, renal function and body mass index. Finally, both NT-proBNP and the albumin gradient correctly identified more than 80% of those cardiac effusions misclassified as exudates by standard criteria. CONCLUSIONS: MR-proANP is as valuable a diagnostic tool as NT-proBNP for diagnosing or excluding HF as the cause of pleural effusion.
BACKGROUND AND OBJECTIVE: The purpose of this study was to compare the diagnostic utility of pleural fluid N-terminal pro-B-type natriuretic peptide (NT-proBNP), midregion pro-atrial natriuretic peptide (MR-proANP) and midregion pro-adrenomedullin (MR-proADM) for discriminating heart failure (HF)-associated effusions. METHODS: NT-proBNP, MR-proANP and MR-proADM were measured by commercially available methodologies in the pleural fluid of a retrospective cohort of 185 consecutive patients with pleural effusions, of whom 95 had acute decompensated HF. Receiver-operating characteristic and area under the curve (AUC) analyses allowed comparisons of the discriminative properties of these biomarkers to be made at their optimal cut-off points. RESULTS: The diagnostic accuracy of NT-proBNP and MR-proANP for HF as quantified by the AUC was 0.935 and 0.918, respectively, whereas MR-proADM was of limited value (AUC = 0.62). A pleural fluid MR-proANP >260 pmol/L or NT-proBNP >1700 pg/mL argues for HF (likelihood ratio (LR) positive >5), while levels below these cut-off values significantly decrease the probability of having the disease (respective LR negative 0.19 and 0.10). The optimal cut-off points for natriuretic peptides were influenced by age, renal function and body mass index. Finally, both NT-proBNP and the albumin gradient correctly identified more than 80% of those cardiac effusions misclassified as exudates by standard criteria. CONCLUSIONS: MR-proANP is as valuable a diagnostic tool as NT-proBNP for diagnosing or excluding HF as the cause of pleural effusion.