Background: Adipokines could pose a link between adiposity, systemic inflammation and metabolic disease risk. However, it is unclear whether representative biomarkers are methodologically suitable for use in human obesity research.
Methods: We assessed the intra-individual reproducibility of selected adipokines in a sample of 207, apparently healthy, participants with available biosample collections over a 4-month period. Concentrations of the following adipokines were measured at each sampling time point: fatty-acid binding protein-4 (FABP-4), lipocalin-2, monocyte chemoattractant protein 1 (MCP-1), procalcitonin, progranulin, vaspin and visfatin/Nampt. We calculated intraclass correlation coefficients (ICC) and examined Bland-Altman plots.
Results: The analyses suggested an overall good to excellent biomarker reproducibility over 4 months: FABP-4: ICC=0.73 (95% confidence interval: 0.65, 0.78), lipocalin-2: 0.64 (0.55, 0.71), MCP-1: 0.85 (0.81; 0.89), procalcitonin: 0.78 (0.72, 0.83), progranulin: 0.59 (0.50, 0.68) and vaspin: 0.86 (0.82, 0.89). A good agreement of the repeated measurements was further supported by the Bland-Altman plots. No substantial differences in biomarker performance according to adiposity status could be observed. Reliability of visfatin/Nampt could not be assessed due to a high number of measurements below the lower limit of detection.
Conclusion: Results suggest that single measurements of the evaluated adipokines could be used in population-based studies aimed to assess links between obesity, inflammation and metabolic diseases.