Importance of early combined N-terminal pro-brain natriuretic peptide and cardiac troponin T measurements for long-term risk stratification of patients with decompensated heart failure.

摘要

BACKGROUND: Markers of myocardial necrosis and natriuretic peptides are risk predictors in decompensated heart failure (DHF). We prospectively studied the optimal timing of combined cardiac troponin T (cTnT) and N-terminal pro-brain natriuretic peptide (NT-proBNP) measurements for long-term risk stratification. METHODS: cTnT and NT-proBNP were measured upon admission, and before discharge in 76 patients hospitalized for DHF (mean age 62.3 +/- 15 years; 71% men). RESULTS: During a mean follow-up of 252 +/- 120 days, 39.5% of patients died or were re-hospitalized for DHF. From receiver-operator-characteristic (ROC) curves, the selected cut-off values for cTnT and NT-proBNP were 0.026 ng/ml and 3,700 pg/ml on admission, and 0.030 ng/ml and 3,200 pg/ml, respectively, at discharge. Depending upon measurements above vs below cut-off, the population was distributed on admission and before discharge for three groups: both negative (24% and 30% of patients); one positive (43% and 42%); and both positive (33% and 28%). For the admission groups, the 1-year DHF-free re-hospitalization survival rates were 85%, 60% and 34%, respectively (p = 0.0047). One-year survival rates for DHF-free re-hospitalization were 63%, 71% and 26% (p = 0.0029), respectively, for the discharge groups. In the Cox proportional hazards model, systolic blood pressure (hazard ratio [HR]: 0.98; 95% confidence interval [CI]: 0.96 to 0.99), heart rate (HR: 0.97; 95% CI: 0.94 to 0.98), one positive biomarker on admission (HR: 10.5; 95% CI: 1.3 to 83.7) and two positive biomarkers on admission (HR: 13.9; 95% CI: 1.8 to 98.5) were independent predictors of long-term outcomes. However, NT-proBNP on admission was the most important predictor of long-term prognosis (HR: 5.1; 95% CI: 2.3 to 12.2). CONCLUSIONS: The combined measurements of cTnT and NT-proBNP on hospital admission were more reliable than their measurements before discharge in the long-term risk stratification of DHF. A single positive measurement on admission predicted a poor long-term outcome.