ABSTRACT
AIMS: Guidelines support the role of B-type natriuretic peptide (BNP) and amino-terminal pro-BNP (NT-proBNP) for risk stratification of patients in programmes to prevent heart failure (HF). Although biologically formed in a 1:1 ratio, the ratio of NT-proBNP to BNP exhibits wide inter-individual variability. A report on an Asian population suggests that molar NT-proBNP/BNP ratio is associated with incident HF. This study aims to determine whether routine, simultaneous evaluation of both BNP and NT-proBNP is warranted in a European, Caucasian population. METHODS AND RESULTS: We determined BNP and NT-proBNP levels for 782 Stage A/B HF patients in the STOP-HF programme. The clinical, echocardiographic, and biochemical associates of molar NT-proBNP/BNP ratio were analysed. The primary endpoint was the adjusted association of baseline molar NT-proBNP/BNP ratio with new-onset HF and/or progression of left ventricular dysfunction (LVD). We estimated the C-statistic, integrated discrimination improvement, and the category-free net reclassification improvement metric for the addition of molar NT-proBNP/BNP ratio to adjusted models. The median age was 66.6 years [interquartile range (IQR) 59.5-73.1], 371 (47.4%) were female, and median molar NT-proBNP/BNP ratio was 1.91 (IQR 1.37-2.93). Estimated glomerular filtration rate, systolic blood pressure, left ventricular mass index, and heart rate were associated with NT-proBNP/BNP ratio in a linear regression model (all P < 0.05). Over a median follow-up period of 5 years (IQR 3.4-6.8), 247 (31.5%) patients developed HF or progression of LVD. Log-transformed NT-proBNP/BNP ratio is inversely associated with HF and LVD risk when adjusted for age, gender, diabetes, hypertension, vascular disease, obesity, heart rate, number of years of follow-up, estimated glomerular filtration rate, and baseline NT-proBNP (odds ratio 0.71, 95% confidence interval 0.55-0.91; P = 0.008). However, molar NT-proBNP/BNP ratio did not increase the C-statistic (Δ -0.01) and net reclassification improvement (0.0035) for prediction of HF and LVD compared with NT-proBNP or BNP alone. Substitution of NT-proBNP for BNP in the multivariable model eliminated the association with HF and LVD risk. CONCLUSIONS: This study characterized, for the first time in a Caucasian Stage A/B HF population, the relationship between NT-proBNP/BNP ratio and biological factors and demonstrated an inverse relationship with the future development of HF and LVD. However, this study does not support routine simultaneous BNP and NT-proBNP measurement in HF prevention programmes amongst European, Caucasian patients.
