RESUMO
BACKGROUND AND OBJECTIVE: The N-terminal brain natriuretic peptide (NT-proBNP) is a new tool for the diagnosis of patients with evidence of heart failure (HF). The diagnosis of HF is improbable in patients with a level of NT-proBNP<300 pg/ml. Our objective is to determine the validity of the cut-off points proposed by literature (inclusion/exclusion) regarding the age for NT-proBNP in patients admitted to our service. PATIENTS AND METHOD: We gathered consecutively the data of 76 patients admitted to our service with the diagnosis of HF (n=37) and with other diseases (n=39), taking a sample of plasma to determine NT-proBNP. We studied the sensitivity, specificity and predictive values of this test, as well as the relation between the levels of that marker and the clinical and echocardiographic variables. RESULTS: The ventricular diameter was larger and the ejection fraction was lower in patients with HF. Sensitivity for the diagnosis of exclusion reached a value of 97% and the negative predictive value was 94%. CONCLUSIONS: The cut-off point proposed for exclusion (300 pg/ml) shows high sensitivity and negative predictive value in the diagnosis of HF in patients admitted to our service.