| Objective Plasma N-terminal pro-brain natriuretic peptide (NT-proBNP) is a strong biochemical marker of chronic heart failure (CHF). Due to renal arterio-venous clearance of NT-proBNP and the correlation of plasma concentrations with renal function,we hypothesized that NT-proBNP may have potential as a urinary marker. But the precise mechanism explaining the increased NT-proBNP concentrations among patients with concomitant CHF and kidney dysfunction is not fully understood. The objective of this study was to assess the impact of kidney dysfunction on simultaneous measures of plasma and urinary NT-proBNP in an unselected cohort of patients with CHF,and further to identify the diagnostic value of plasma and urinary NT-proBNP for detecting CHF.Method We studied 106 patients (age, 66±14 years; 45% men) with CHF and 22 sex-matched healthy control subjects. NYHA classification was used to evaluate the clinical conditions of patients.Left ventricular ejection fraction(LVEF) was measured by echocardiography,and the concentration of plasma and urinary NT-proBNP was detemined by enzyme-linked immunosorbent assay.Estimated glomemlar filtration rate(eGFR)was estimated by MDRD formula.Descriptive variables are presented as mean+SD for all normally distributed variables or median [inter-quartile range (IQR)] for nonparametric variables. Non-parametric variables were log-transformed to normalize variances before analyses.Contrast in different group was done with T-test and One-way ANOVA. The independent effect of clinical variables on plasma and urinary NT-proBNP concentrations was investigated by using multiple linear regression analyses.the value of the plasma and urinary NT-proBNP to diagnose CHF was estimate by the receiver operating characteristic(ROC) curve.Data analysis was performed using SPSS version 19.0. Statistical significance was assumed at P< 0.05.Results In patients with CHF (n=106), plasma NT-proBNP concentrations were higher than simultaneous urinary NT-proBNP concentrations (3149.6 [1170.2–7531.2] pg `mL vs 51.2 [20.5–164.9]pg `mL; P<0.001). The mean eGFR for all patients was 81.9±28.8 mL·min-1·1.73 m-2. Both plasma and urinary NT-proBNP concentrations decreased across eGFR categories (P<0.001). Correlations between eGFR and NT-proBNP concentrations were also significant,although less robust (r=-0.459 for plasma NT-proBNP and r=-0.365 for urinary NT-proBNP, P<0.001 for both).There were no differences in the ratio of NT-proBNP across eGFR categories(P=0.545). The NT-proBNP ratio was not significantly correlated with eGFR (r=-0.053,P=0.461).Although reduced eGFR was associated with increased plasma and urinary NT-proBNP concentration in univariate analyses.However, after multivariate adjustment, eGFRwas found to be an independent predictor of plasma (but not urinary) NT-proBNP concentration (β=-0.330,P<0.001). Indeed, plasma NT-proBNP was the main independent determinant of its urinary concentration (β=0.409, P<0.001).Urinary and plasma NT-proBNP concentrations were positively correlated (r=0.605, P<0.001).Plasma and urinary NT-proBNP showed a stepwise and significant increase with deteriorating LVEF and a negative correlation with LVEF (r=-0.582 vs r=-0.528; P<0.001).Upon ROC curve analysis, plasma NT-proBNP detected CHF with a sensitivity of 90.6% and a specificity of 86.4% at a cutpoint of 705.1pg/mL [area under the curves (AUC) 0.956].Urinary NT-proBNP detected CHF with a sensitivity of 88.7% and a specificity of 86.4% at a cutpoint of 10.2 pg/mL (AUC 0.924).Conclusion In patients with CHF and concomitant kidney dysfunction, the increased circulating NT-proBNP may be mainly related to increased cardiac secretion and not decreased renal clearance. NT-proBNP concentrations were markedly lower in the urine than in the plasma.However, urinary NT-proBNP levels increased stepwise with the severity of CHF,and therefore yielded satisfactory predictive values for the detection of CHF. Measurement of urinary NT-proBNP is a novel, promising, and simple method for the biochemical detection of heart failure.In patients with CHF and concomitant kidney dysfunction, plasma and urinary NT-proBNP satisfactory predictive values for the detection of CHF. |