Premium
FGF‐23 and Osteoprotegerin but not Fetuin‐A are associated with death and enhance risk prediction in non‐dialysis chronic kidney disease stages 3–5.
Author(s) -
Alderson Helen V,
Ritchie James P,
Middleton Rachel,
Larsson Anders,
Larsson Tobias E.,
Kalra Philip A
Publication year - 2016
Publication title -
nephrology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 61
eISSN - 1440-1797
pISSN - 1320-5358
DOI - 10.1111/nep.12664
Subject(s) - medicine , osteoprotegerin , fibroblast growth factor 23 , kidney disease , hazard ratio , dialysis , interquartile range , proportional hazards model , risk factor , confidence interval , parathyroid hormone , receptor , activator (genetics) , calcium
Aim Numerous biomarkers have been shown to associate with clinical endpoints in chronic kidney disease (CKD). There is limited evidence whether biomarkers improve risk prediction in relation to clinical outcomes. Our study investigates whether a small suite of key chronic kidney disease‐mineral and bone disorder biomarkers could be used to enhance risk assessment in CKD. Methods Fetuin‐A, fibroblast growth factor‐23 and osteoprotegerin were measured on baseline plasma samples from 463 patients recruited to the Chronic Renal Insufficiency Standards Implementation Study. The biomarkers were analysed in relation to progression to end stage kidney disease, death and major cardiovascular events. Results Over a median follow up of 46 months (interquartile range 21–69), fibroblast growth factor‐23 was associated with risk for renal replacement therapy (hazard ratio (HR) 1.35, P = 0.05, 95% confidence interval (CI) 1.001–1.820), cardiovascular events (HR 1.74 P < 0.001, 95% CI 1.303–1.305) and death (HR 1.4 P = 0.005, 95% CI 1.109–1.767). Osteoprotegerin was associated with risk for death (HR 1.06, P = 0.03, 95% CI 1.006–1.117). There was no clear association between Fetuin‐A and any of the clinical endpoints. The addition of biomarkers to risk models led to marginal improvement in model discrimination and reclassification. Conclusion Biomarkers are often associated with clinical endpoints, and we observed such associations in our study of patients with advanced CKD. However, the markers analysed in our study were of limited benefit in improving the prediction of these outcomes. Any extra information biomarkers may provide to improve risk prediction in clinical practice needs to be carefully balanced against the potential cost of these tools.