Natriuretic peptides as markers of mild forms of left ventricular dysfunction: effects of assays on diagnostic performance of markers.

BACKGROUND We compared the performance of different natriuretic peptides to diagnose mild forms of left ventricular dysfunction (LVD) and investigated the influence of measuring B-type natriuretic peptide (BNP) and N-terminal proBNP (NT-proBNP) with different assays on the diagnostic performance of these markers. METHODS We measured BNP (Triage BNP), NT-proBNP (Biomedica), and N-terminal pro-A-type natriuretic peptide (NT-proANP; Biomedica) in 130 consecutive patients (age range, 28-83 years) with clinically suspected mild LVD. In patients with sufficient sample volume, we measured BNP and NT-proBNP with additional assays (Shionoria and Roche, respectively). RESULTS For identifying patients with mild systolic LVD, BNP and NT-proBNP were the best markers, with mean (95% confidence interval) areas under the curves (AUC) of 0.78 (0.63-0.89) and 0.75 (0.58-0.87), respectively. However, the diagnostic performance of NT-proANP [AUC, 0.64 (0.48-0.77)] was significantly worse than that of BNP (P = 0.014). Both BNP assays (Triage and Shionoria) and both NT-proBNP assays (Biomedica and Roche) performed equally well for the diagnosis of systolic LVD despite the poor agreement between NT-proBNP assays. In patients with isolated diastolic LVD, the diagnostic performance of the Triage BNP [AUC, 0.70 (0.56-0.81)] was significantly better (P = 0.006) than that of Biomedica NT-proBNP [0.49 (0.34-0.65)]. Furthermore, the performance of the Biomedica NT-proBNP assay was significantly worse (P = 0.03) than that of the Roche NT-proBNP assay for diagnosis of isolated diastolic LVD. CONCLUSIONS The performance of BNP for the diagnosis of systolic or diastolic LVD is not affected by the assay used, whereas the performance of NT-proBNP for the diagnosis of isolated diastolic LVD is assay dependent.