Background: The prognostic value of natriuretic peptides in heart failure (HF) is well established. The objective of this study was to evaluate the role of NT-proBNP in predicting outcome in decompensated HF.
Methods: Patients admitted with decompensated HF to our Internal Medicine Department between November 2002 and April 2004 with at least two measurements of NT-proBNP (within 24 hours of admission and on discharge) were analyzed. Patients discharged alive were followed for up to 6 months. The primary endpoint was death or readmission.
Results: We included 304 patients (72.7+/-11.6 years of age, 53.9% female, 49.3% ischemic etiology). Echocardiography was performed in 73.7%. Left ventricular systolic function (LVSF) was preserved in 20.7%, mildly to moderately depressed in 32.2% and severely depressed in 20.7%. There was a significant decrease in median NT-proBNP levels during hospitalization (from 7006 to 3796 pg/ml, p<0.001). The patients were classified in three groups according to NT-proBNP variation: 1 decreasing by at least 30% (n=162); 2 - no significant variation (n=95); and 3 - increasing by at least 30% (n=47). The primary endpoint was observed in 43% of the patients. In univariate analysis, variables predictive of outcome were: NT-proBNP at discharge (> median: HR=2.72; 95% CI=1.89-3.92): variation in NT-proBNP levels during hospitalization (group 2 vs. group 1 - HR=2.28; 95% CI=1.52-3.42; group 3 vs. group 1 - HR=4.82; 95% CI=3.11-7.49); renal failure (creatinine >2 mg/dL - HR=1.65; 95% CI=1.07-2.53); and treatment with ACE-Is (HR=0.59; 95% CI=0.39-0.89). After adjustment for NYHA class at discharge, pulse pressure, LVSF, renal function and hemoglobin, only NT-proBNP at discharge and NT-proBNP variation remained independent predictors of prognosis (NT-proBNP at discharge > median: HR=2.02; 95% CI=1.28-3.2; NT-proBNP variation: group 2 vs. group 1 - HR=2.24; 95% CI=1.37-3.66; group 3 vs. group 1 - HR=3.85; 95% CI=2.24-6.63).
Conclusion: Our results extend previous reports on the value of NT-proBNP in predicting outcome after discharge in patients hospitalized due to decompensated HF, and demonstrate its potential usefulness and applicability in clinical practice.