Abstract
Hemoglobin and amino-terminal pro-brain natriuretic peptide (NT-proBNP) are both independent predictors of mortality in patients with chronic HF. Their combined predictive power for mortality in the setting of acute HF is uncertain.
In an international prospective cohort design, we evaluated the relationships between hemoglobin, NT-proBNP, and 60-day mortality in 690 patients with acute HF.
The median hemoglobin for the entire cohort was 13.0 g/dL (interquartile range 11.6-14.3). The WHO criterion for anemia was met by 44% (n=305). The 60-day mortality rate for anemic patients was 16.4% vs. 8.8% in non-anemic patients (p<0.001). Anemia was an independent predictor of short-term mortality (OR=1.72, 95% CI=1.05-2.80, p=0.03), as was a NT-proBNP concentration >5180 pg/mL (OR=2.32, 95% CI=1.36-3.94 p=0.002). Consideration of four risk groups: not anemic/low NT-proBNP (reference group, n=220), anemic/low NT-proBNP (n=152), not anemic/high NT-proBNP (n=165), and anemic/high NT-proBNP (n=153) revealed respective 60-day mortality rates of 5.0% (referent), 9.2% (OR=1.93, 95% CI=0.85-4.36; p=0.12), 13.9% (OR=3.07, 95% CI=1.45-6.50, p=0.003), and 23.5% (OR=5.84, 95% CI=2.87-11.89, p<0.001).
Anemia was common in this cohort of subjects with acute HF and was related to adverse short-term outcome. Integrated use of hemoglobin and NT-proBNP measurements provides powerful additive information and is superior to the use of either in isolation.