Consecutive hospitalized patients with ADHF were prospectively studied. Blood and urine samples were simultaneously collected on hospital arrival to determine NT-proBNP concentrations. Clinical follow-up was obtained, and the occurrence of mortality and heart failure hospitalization was registered.
The study included 138 patients (median, 74 years [interquartile range, 67-80]; 54 % men). During a median follow-up period of 387 days [interquartile range, 161-559], 65 patients (47 % ) suffered adverse clinical events. Plasma NT-proBNP concentration was higher among patients who presented adverse events (4561 pg/mL [2191-8631] vs 2906 pg/mL [1643-5823]; P = .03), whereas urinary NT-proBNP was similar in both groups (P = .62). After multivariable Cox regression analyses, plasma NT-proBNP concentration was associated with a higher risk of adverse events, whether considered continuously (per 100 pg/mL; hazard ratio [HR] = 1.004; 95 % confidence interval [CI], 1.001-1.007; P = .003) or categorically (≥3345 pg/mL; HR = 2.35; 95 % CI, 1.41-3.93; P = .001). In contrast, urinary NT-proBNP concentration was not associated with adverse outcomes.
Plasma NT-proBNP concentration is superior to urinary NT-proBNP concentration for the prediction of adverse clinical outcomes among unselected patients with ADHF.