Prognostic usefulness of anemia and N-terminal pro-brain natriuretic peptide in outpatients with systolic heart failure

N-terminal pro-brain natriuretic peptide (NT-pro-BNP) and anemia are predictors of outcome in systolic heart failure. It is currently unclear how these 2 markers interact in particular with regard to the prognostic information carried by each risk marker. We therefore tested the hypothesis that anemia (World Health Organization criteria, hemoglobin levels <7.5 mmol/L for women and <8.0 mmol/L for men) and NT-pro-BNP are associated and evaluated how a possible association affects the prognostic value of each risk marker. Clinical data from 345 patients with systolic heart failure were obtained prospectively at the baseline visit to our heart failure clinic (inclusion criterion left ventricular ejection fraction <0.45, no exclusion criteria). Follow-up was 30 months (median), and 70 events (mortality) occurred. Prevalence of anemia was 27%. In a multivariate logistic regression model, anemia (p = 0.041) was closely associated with NT-pro-BNP levels above the median (1,381 pg/ml) after adjustment for traditional confounders (left ventricular ejection fraction, age, body mass index, atrial fibrillation, chronic kidney disease). In an adjusted Cox proportional hazard model, the 2 parameters were associated with mortality after adjustment for traditional confounders (hazard ratio for anemia 1.73, 95% confidence interval 1.06 to 2.83, p = 0.029; hazard ratio for NT-pro-BNP >1,381 pg/ml 2.68, 95% confidence interval 1.58 to 4.66, p <0.001). Patients with anemia and high NT-pro-BNP levels had a fivefold increased risk for mortality (hazard ratio 4.77, 95% confidence interval 2.47 to 9.18, p <0.001). In conclusion, anemia is closely associated with NT-pro-BNP in patients with systolic heart failure, and anemia and NT-pro-BNP carry independent prognostic information. Patients with anemia and high levels of NT-pro-BNP have a markedly increased mortality risk.