Abstract
AIMS: To test the prognostic value of emerging biomarkers in the Reduction of Events by Darbepoetin Alfa in Heart Failure (RED-HF) trial. METHODS AND RESULTS: Circulating cardiac [N-terminal pro-B-type natriuretic peptide (NT-proBNP), and high-sensitivity troponin T (hsTnT)], neurohumoral [mid-regional pro-adrenomedullin (MR-proADM) and copeptin], renal (cystatin C), and inflammatory [high-sensitivity C-reactive protein (hsCRP)] biomarkers were measured at randomization in 1853 participants with complete data. The relationship between these biomarkers and the primary composite endpoint of heart failure hospitalization or cardiovascular death over 28 months of follow-up (n = 834) was evaluated using Cox proportional hazards regression, the c-statistic and the net reclassification index (NRI). After adjustment, the hazard ratio (HR) for the composite outcome in the top tertile of the distribution compared to the lowest tertile for each biomarker was: NT-proBNP 3.96 (95% CI 3.16-4.98), hsTnT 3.09 (95% CI 2.47-3.88), MR-proADM 2.28 (95% CI 1.83-2.84), copeptin 1.66 (95% CI 1.35-2.04), cystatin C 1.92 (95% CI 1.55-2.37), and hsCRP 1.51 (95% CI 1.27-1.80). A basic clinical prediction model was improved on addition of each biomarker individually, most strongly by NT-proBNP (NRI +62.3%, P < 0.001), but thereafter was only improved marginally by addition of hsTnT (NRI +33.1%, P = 0.004). Further addition of biomarkers did not improve discrimination further. Findings were similar for all-cause mortality. CONCLUSION: Once NT-proBNP is included, only hsTnT moderately further improved risk stratification in this group of chronic heart failure with reduced ejection fraction patients with moderate anaemia. NT-proBNP and hsTnT far outperform other emerging biomarkers in prediction of adverse outcome.