BNP-guided therapy optimizes the timing of discharge and the medium term risk stratification in patients admitted for congestive heart failure

BACKGROUND: Congestive heart failure (HF) is one of the most important cause of hospitalizations and is associated with high cost. Despite a consistent body of data demonstrating the benefits of drug therapy in HF, persistently high rates of readmission, especially within six months of discharge, continue to be documented. Neurohormonal activation characterizes the disease; plasma brain natriuretic peptide (BNP), is correlated with the severity of left ventricular dysfunction and relates to outcome. OBJECTIVE: The aim of the study was to evaluate if plasma levels of BNP would provide an index to guide drug treatment and to predict medium-term prognosis in HF patients (pts) after hospital discharge. METHODS AND RESULTS: We evaluated 200 consecutive pts (age 77 +/- 10 (35-96) years, 49% male versus 51% female) hospitalized for HF (DRG 127). Standard echocardiography was performed and left ventricular systolic/diastolic function was assessed; plasma BNP levels were measured with a rapid point-of-care assay (Triage BNP Test, Biosite Inc, San Diego, CA) on days 1 and after initial treatment. Using a cut-off of 240 pg/ml and/or changes in plasma BNP (days 2-3 after admission), 2 groups were identified: the low BNP group-responders (n = 68, BNP < 240 pg/ml and/or > or = 30% reduction) and the high BNP group-non responders (n = 132, BNP > or = 240 pg/ml and/or < 30% reduction). The high BNP group showed a different pattern of clinical variables according to the severity of the disease New York Heart Association (NYHA) functional class, left ventricular ejection fraction, ischemic etiology and age. A sustained elevation of plasma BNP (> 240 pg/mL) indicated the presence of a clinical unstable condition requiring further intervention whereas pts with low BNP values were discharged after 24 hours. During a mean follow-up period of 3 months, there were 62 cardiac events, including 15 cardiac deaths, 22 readmissions for worsening heart failure and 25 clinical decompensation requiring diuretic treatment. The incidence of clinical events was significantly greater in pts with higher levels of BNP (admission and discharge) than in those with lower levels (42% vs. 10%) and plasma values > 500 pg/ml identified a subgroup at high risk of death. CONCLUSIONS: The influence of BNP in the clinical course and prognosis of patients hospitalized for HF has not been studied. After initial treatment pts need to be risk stratified by means of the BNP test, to guide further management and to identify subjects with poor prognosis. An aggressive therapeutic and follow-up strategy may be justified for pts with high BNP levels and/or no changes after hospital admission for worsening HF. The changes in plasma BNP level at discharge were significantly related to cardiac events.