Mineral metabolites and CKD progression in African Americans

Abstract

CKD progresses more rapidly to ESRD among African Americans compared with Caucasians. Disordered mineral metabolism is more severe among African Americans with CKD, which might partially explain the accelerated progression of their kidney disease. Here, using data from the African American Study of Kidney Disease and Hypertension, we evaluated longitudinal changes in serum levels of fibroblast growth factor-23 (FGF23), parathyroid hormone (PTH), phosphate, and 25-hydroxyvitamin D in a subset of 420 participants followed for a median of 4 years. We also examined the association of baseline levels of mineral metabolites with risk for ESRD or death in 809 participants. FGF23, PTH, and phosphate levels rose over time; participants with faster rates of decline in measured GFR had the greatest increases in these parameters (P<0.01 for each). Higher baseline levels of FGF23, PTH, and phosphate each associated with increased risk for ESRD or death independent of GFR. FGF23 exhibited a dose-response relationship with outcomes (HR=1.30 per doubling, 95% CI=1.15-1.47; HR=2.24 for highest compared with lowest quartile, 95% CI=1.39-3.60), whereas PTH and phosphate showed nonlinear relationships. Vitamin D insufficiency (<30 ng/ml) was present in 95% of participants, but lower levels did not independently associate with outcomes. Using death-censored ESRD as the outcome produced qualitatively similar results. In conclusion, abnormalities of mineral metabolism worsen with progressive CKD and associate with higher risk for ESRD among African Americans with hypertensive nephrosclerosis.

Figure 1.

Association between mineral metabolites and adjusted hazard of ESRD or death. Models were performed using restricted cubic splines with knots at the 10th, 50th, and 90th percentiles. (A) Natural log-transformed (ln) FGF23, (B) ln PTH, and (C) serum phosphate. Solid line represents estimated HR, shaded area represents the 95% CI, and histogram represents the distribution of the mineral metabolite in the study population. Models are adjusted for age, sex, income, prior heart disease, smoking, UPCR, GFR, randomized treatment groups, categories of body mass index, and serum albumin, and they are clustered by clinical center.

Figure 2.

Incidence of ESRD or death by the number of mineral metabolism abnormalities versus stage of CKD. Black squares represent incidence rate estimates, and vertical lines represent the 95% CIs calculated using a Poisson variance distribution.