Background: We investigate whether the urinary proteome refines the diagnosis of renal dysfunction, which affects over 10% of the adult population.
Methods: We measured serum creatinine, estimated glomerular filtration rate (eGFR) and 24-h albuminuria in 797 people randomly recruited from a population. We applied capillary electrophoresis coupled with mass spectrometry to measure multi-dimensional urinary proteomic classifiers developed for renal dysfunction (CKD273) or left ventricular dysfunction (HF1 and HF2). Renal function was followed up in 621 participants and the incidence of cardiovascular events in the whole study population.
Results: In multivariable-adjusted cross-sectional analyses, higher biomarker levels analysed separately or combined by principal component analysis into a single factor (SF), correlated (P ≤ 0.010) with worse renal function. Over 4.8 years, higher HF1 and SF predicted (P ≤ 0.014) lowering of eGFR; higher HF2 predicted (P ≤ 0.049) increase in serum creatinine and decrease eGFR. HF1, HF2 and SF predicted progression from CKD Stages 2 or ≤2 to Stage ≥3, with risk estimates for a 1-SD increment in the urinary biomarkers ranging from 38 to 71% (P ≤ 0.039). HF1, HF2 and SF yielded a net reclassification improvement of 31-51% (P ≤ 0.029). Over 6.1 years, 47 cardiovascular events occurred. HF2 and SF, independent of baseline eGFR, 24-h albuminuria and other covariables were significant predictors of cardiovascular complications with risk estimates for 1-SD increases ranging from 32 to 41% (P ≤ 0.047).
Conclusions: The urinary proteome refines the diagnosis of existing or progressing renal dysfunction and predicts cardiovascular complications.
Keywords: chronic kidney disease; eGFR; population science; renal function; urinary proteomics.
© The Author 2014. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.