LAUSR

BACKGROUND Circulating concentrations of N-terminal fragment of the prohormone of brain natriuretic peptide (NT-proBNP) are influenced by age and common age-related comorbidities, such as renal dysfunction. Therefore, utility of NT-proBNP for prediction of prognosis in the aged has been questioned. We aimed to investigate the prognostic value of NT-proBNP across age classes in a cohort of patients with chronic systolic HF. METHODS AND RESULTS We enrolled 2364 consecutive outpatients with HF and left ventricular ejection fraction ≤50%. Patients were classified according to age quartiles, and a very elderly subgroup was identified, aged ≥85 years. After baseline assessment (including NT-proBNP testing), patients were followed-up for the composite of cardiovascular death, heart transplantation or ventricular assistance device implantation (primary outcome) and for all-cause death (secondary outcome). Patients in the fourth quartile (Q4, age ≥ 77 years, n = 638) and in the very elderly subgroup (age ≥ 85 years, n = 153), had higher NT-proBNP (p < .001 vs Q1). NT-proBNP was independently associated with primary and secondary outcome at 1- and 5-years follow-up in the whole population, as well as in Q4 and in the very elderly subgroup (all p < .05). Compared to the whole population, Q4 and very elderly had higher NT-proBNP cut-off for prediction of 1-year primary (4188 and 9729 ng/l, respectively vs 3710 ng/l) and secondary outcome (4296 and 7634 ng/l, respectively vs 3056 ng/l). CONCLUSIONS NT-proBNP predicts mortality in elderly and very elderly patients with chronic systolic HF, both at mid- and long-term follow-up. Higher NT-proBNP prognostic cut-off should be considered in the aged HF population.