Changes in Natriuretic Peptide Levels and Subsequent Kidney Function Decline in SPRINT

Abstract

Rationale & ObjectiveNovel approaches to assessing kidney disease risk during hypertension treatment are needed given the uncertainty in how intensive blood pressure (BP) lowering impacts kidney outcomes. We determined whether longitudinal N-terminal pro-B-type natriuretic peptide (NT-proBNP) measurements during hypertension treatment are associated with kidney function decline. Study DesignProspective observational study. Setting& Participants: 8005 Systolic Blood Pressure Intervention Trial (SPRINT) participants with NT-proBNP measurements at baseline and 1 year. Exposure1-year change in NT-proBNP categorized as ≥25% decrease, ≥25% increase, and <25% change (stable). OutcomeAnnualized change in estimated glomerular filtration rate (eGFR) and ≥30% decline in eGFR. Analytical ApproachLinear mixed-effect and logistic regression models were used to evaluate the association of change in NT-proBNP with subsequent annualized change in eGFR and ≥30% decline in eGFR, respectively. Analyses stratified by baseline chronic kidney disease (CKD) status. ResultsCompared with stable 1-year NT-proBNP levels, a ≥25% decrease in NT-proBNP was associated with slower decline in eGFR in those with CKD (adjusted difference [% per year] =1.09 [95% CI, 0.35-1.83]) and without CKD (adjusted difference [% per year] =0.51 [95% CI, 0.21-0.81]; P for interaction =0.4). Meanwhile, ≥25% increase in NT-proBNP in those with CKD was associated with faster decline in eGFR (adjusted difference [% per year] = -1.04 [95% CI, -1.72 to -0.36]) and risk of ≥30% decline in eGFR (adjusted odds ratio: 1.44 [95% CI, 1.06-1.96]); associations were stronger in those with CKD versus without CKD (P for interactions =0.01 and <0.001, respectively). Relationships were similar irrespective of the randomized BP arm in SPRINT (P for interactions >0.2). LimitationsPersons with diabetes and proteinuria > 1 g/d were excluded. ConclusionsChanges in NT-proBNP during BP treatment are independently associated with subsequent kidney function decline, particularly in people with CKD. Studies should assess if dynamic NT-proBNP may be useful in monitoring kidney risk during hypertension treatment.