Clinical factors associated with B-type natriuretic peptide reduction rate during hospitalization in patients with acute heart failure
T Kawata, A IkedaAbstract
Background
Although the results of individual studies have been inconsistent regarding whether plasma B-type natriuretic peptide (BNP)-guided therapy improves prognosis in the treatment of heart failure (HF), a recent meta-analysis suggests that it reduces mortality and re-hospitalization for HF. On the other hand, clinical factors associated with BNP reduction rate during hospitalization in patients with acute HF are not fully elucidated.
Purpose
To explore which clinical factors are associated with BNP reduction rate during hospitalization in patients with acute HF.
Methods
We retrospectively enrolled 238 patients (median age, 84 years; 135 male) admitted to our hospital for the treatment of acute decompensated HF. Patients with concurrent obvious infections (e.g., respiratory infections or urinary tract infections) were excluded from this study. Vital signs, clinical characteristics including laboratory and echocardiographic data were obtained on admission. BNP reduction rate (%) was defined as: 100 × (BNP on admission - BNP at discharge) / BNP on admission. We investigated association between BNP reduction rate and each clinical factors obtained on admission, including factors known to influence BNP concentration such as body mass index, hemoglobin level and estimated glomerular filtration rate.
Results
The median BNP concentration on admission and at discharge was 700pg/ml and 216pg/ml, respectively, and the median reduction rate in BNP concentration during hospitalization was 64.2%. In univariate regression analysis, younger age, higher systolic blood pressure (BP), higher hemoglobin level, larger left ventricular (LV) mass index, lower LV ejection fraction and lower tricuspid regurgitation velocity were associated with large BNP reduction rate. Moreover, presence of atrial fibrillation (AF) inversely associated with large BNP reduction rate. Multivariate regression analysis including body mass index and estimated glomerular filtration rate revealed that systolic BP (β = 0.18, P = 0.0066), LV mass index (β = 0.15, P = 0.039) and presence of AF (β = -0.13, P = 0.049) were independently associated with BNP reduction rate.
Conclusion
Both higher systolic BP on admission and larger LV mass index were independently associated with a larger reduction in BNP concentration, whereas the presence of AF was independently associated with a smaller reduction in BNP concentration. These results may contribute to understanding the characteristics of HF patients who are prone to or resistant to BNP reduction during hospitalization.