Kidney injury biomarkers and urinary creatinine variability in nominally healthy adults

Biomarkers. 2015;20(6-7):436-52. doi: 10.3109/1354750X.2015.1094136.

Abstract

Environmental exposure diagnostics use creatinine concentrations in urine aliquots as the internal standard for dilution normalization of all other excreted metabolites when urinary excretion rate data are not available. This is a reasonable approach for healthy adults as creatinine is a human metabolite that is continually produced in skeletal muscles and presumably excreted in the urine at a stable rate. However, creatinine also serves as a biomarker for glomerular filtration rate (efficiency) of the kidneys, so undiagnosed kidney function impairment could affect this commonly applied dilution calculation. The United States Environmental Protection Agency (US EPA) has recently conducted a study that collected approximately 2600 urine samples from 50 healthy adults, aged 19-50 years old, in North Carolina in 2009-2011. Urinary ancillary data (creatinine concentration, total void volume, elapsed time between voids), and participant demographic data (race, gender, height, and body weight) were collected. A representative subset of 280 urine samples from 29 participants was assayed using a new kidney injury panel (KIP). In this article, we investigated the relationships of KIP biomarkers within and between subjects and also calculated their interactions with measured creatinine levels. The aims of this work were to document the analytical methods (procedures, sensitivity, stability, etc.), provide summary statistics for the KIP biomarkers in "healthy" adults without diagnosed disease (distribution, fold range, central tendency, variance), and to develop an understanding as to how urinary creatinine level varies with respect to the individual KIP proteins. Results show that new instrumentation and data reduction methods have sufficient sensitivity to measure KIP levels in nominally healthy urine samples, that linear regression between creatinine concentration and urinary excretion explains only about 68% of variability, that KIP markers are poorly correlated with creatinine (r(2) ∼ 0.34), and that statistical outliers of KIP markers are not random, but are clustered within certain subjects. In addition, we interpret these new adverse outcome pathways based in vivo biomarkers for their potential use as intermediary chemicals that may be diagnostic of kidney adverse outcomes to environmental exposure.

Keywords: Computational biology; environmental pollution/ecotoxicology; growth factors/cytokines/inflammatory mediators; immunotoxicity; metabol(n)omics.

MeSH terms

  • Adult
  • Algorithms
  • Biomarkers / urine*
  • Creatinine / urine*
  • Female
  • Humans
  • Kidney Diseases / diagnosis*
  • Kidney Diseases / physiopathology
  • Kidney Diseases / urine*
  • Linear Models
  • Male
  • Middle Aged
  • Models, Biological
  • Reference Values
  • Sensitivity and Specificity
  • United States
  • United States Environmental Protection Agency
  • Young Adult

Substances

  • Biomarkers
  • Creatinine