Abstract Background Chronic kidney disease (CKD) and cardiovascular disease (CVD) exhibit a bidirectional and complex relationship, and share common risk factors and mechanisms of disease development. Thus, important CVD-related biomarkers, such as N-terminal prohormone of B-type natriuretic peptide (NT-proBNP), may serve as a biomarker for CKD development. Purpose We aimed to investigate the cross-sectional and longitudinal associations of NT-proBNP with kidney function and CKD in data from two population-based cohort studies. Methods We included 5854 men and women, aged 25-74 years, from the Cooperative Health Research in the Region of Augsburg (KORA) cohort studies S3 and S4, and their 1-2 subsequent follow-up surveys, with a median follow-up time of 9.4 and 13.4 years, respectively. Kidney function was assessed by 3 estimated glomerular filtration rate (eGFR) assessments, calculated using creatinine (eGFRcr), cystatin C, and combined creatinine and cystatin C. Serum NT-proBNP was logarithmically transformed followed by Z-score standardization and categorized into 4 groups (G1-4), <48.6, 48.6-125, 125-300, and ≥300 pg/ml. In cross-sectional analysis, linear regression and logistic regression were used to estimate β/odds ratio (OR) and their 95% confidence intervals (CIs) of continuous eGFR or prevalent CKD. For longitudinal analysis of repeated eGFR measurements during follow-up (5854 participants, 11870 observations), linear mixed-effects models were used. For incident CKD associations, interval-censored Cox regression models were performed in participants free of CKD at baseline. Results In cross-sectional analyses, higher NT-proBNP levels were associated with lower eGFR and higher prevalence of CKD across all 3 eGFR assessments. For example, results based on eGFRcr showed that the fully-adjusted β (95% CIs) for per 1 SD increase in log-transformed NT-proBNP were -1.03 (-1.42, -0.64) and -7.01 (-8.93, -5.09) 60 ml/min/1.73m2 for G4 compared to G1, respectively. Corresponding OR (95% CIs) were 2.22 (1.81, 2.74) and 13.8 (5.76, 36.4), respectively. In longitudinal analyses, participants with higher baseline NT-proBNP had a faster decline in eGFR during follow-up. The fully-adjusted β (95% CIs) for 5-year eGFRcr decline were -0.67 (-0.87, -0.47) per 1 SD increase, and -0.37 (-0.79, 0.05), -1.53 (-2.18, -0.88), -4.24 (-5.57, -2.90) 60 ml/min/1.73m2/year for G2-4. Higher NT-proBNP levels were associated with higher incident CKD risks. For associations with incident creatinine-based CKD, hazard ratio (95% CIs) were 1.16 (1.00, 1.33) per 1 SD increase and 1.03 (0.77, 1.38), 1.07 (0.75, 1.52), and 1.81 (1.07, 3.04) for G2-4. Conclusions We found associations of higher NT-proBNP with lower kidney function and higher CKD prevalence. Moreover, higher baseline NT-proBNP levels were associated with faster kidney function decline and higher incident CKD risks. Our findings indicate the potential utility of NT-proBNP as a biomarker of kidney health.Figure 1.Cross-sectional associationsFigure 2.Longitudinal associations
Read full abstract