Adding serial N-terminal pro brain natriuretic peptide measurements to optimal clinical management in outpatients with systolic heart failure: a multicentre randomized clinical trial (NorthStar monitoring study)

This study was designed to evaluate a new NT-proBNP monitoring concept in outpatients with systolic heart failure (HF). This was a multicentre, prospective randomized open-label blinded endpoint study. A total of 407 systolic HF patients were allocated to either clinical management (n 208) or clinical management NT-proBNP monitoring (n 199) and followed for 2.5 years. If NT-proBNP increased 30, a clinical checklist was completed and treatment initiated. The patients were matched at randomization and were 73 years old, 25 were females, 85 were NYHA class III, LVEF was 30, and NT-proBNP 1955 pg/mL. NT-proBNP monitoring did not improve outcome, the hazard ratio for the primary composite endpoint (death or a cardiovascular admission) being 0.96 [95 confidence interval (CI) 0.711.29, P 0.766]. NT-proBNP monitoring did not induce a significant change in the pharmacological strategy (P 0.05 for all comparisons). In patients in whom NT-proBNP increased 30 (25 of the patients) during follow-up, a higher frequency of admission (69 vs. 47, P 0.002), a higher number of admission days (14 vs. 5 days, P 0.003) and number of admissions (2 vs. 1, P 0.009), and a lower quality of life (P 0.032) and a poorer functional class (37 vs. 18 in NYHA class IIIIV, P 0.001) were observed. Adding serial measurements of NT-proBNP to optimal clinical management was not associated with a change in pharmacological strategy and did not improve outcome. However, survivors in whom NT-proBNP increased 30 showed a poorer functional class, clinical outcome, and quality of life. www.centerwatch: 173491 (NorthStar).