Background: NT-proBNP is useful for heart failure (HF) diagnosis and prognosis. We examined the value of serial NT-proBNP monitoring to predict outcomes in decompensated HF patients attending a structured HF clinic.
Methods: Patients with decompensation of established optimally treated HF, not requiring emergency hospital admission, were enrolled in the study. Patients received intensive follow-up weekly during 4 weeks and at 3 months in specialist HF clinics. Serial NT-proBNP concentrations were measured at each visit. Primary endpoint was cardiovascular death and hospital admission for HF at 3 months.
Results: Fifty-nine patients were enrolled (60+/-14 years, LVEF 27+/-9%) and 39% had a primary endpoint during follow-up. Baseline NT-proBNP concentration (in ng/L) was 7050+/-6620, and did not differ significantly in patients with and without events (p=0.22). Patients without events showed marked NT-proBNP reduction at week-1 (30% reduction), week-2 (36% reduction), week-3 (34% reduction) and week-4 (37% reduction). By contrast, patients with events showed no significant NT-proBNP reduction during follow-up. Using a general linear model, the relative NT-proBNP reductions (%) at weeks 1-4 were predictors of adverse events (p=0.004, p<0.001, p=0.001 and p=0.03, respectively). In a stepwise multiple Cox regression analysis, NT-proBNP relative reduction (in %) at week 2 was a strong predictor of no events during follow-up (OR 0.79, 95% CI 0.70-0.88, p<0.001).
Conclusions: Serial NT-proBNP monitoring in decompensated HF patients seen in a structured in-hospital HF clinic predicts cardiovascular events during follow-up. NT-proBNP may be useful in an outpatient basis to identify patients at high risk needing more aggressive therapy.