In the mature skeleton approximately 2m bone multicellular units (BMUs) are actively remodelling at any one time. This dynamic process reflects bone turnover. Biochemical markers specific for bone formation and resorption have been proposed as a means to study acute changes in bone turnover in response to, for example, physical activity.

This study sought to (i) measure the variability and (ii) determine the critical difference of plasma and urinary markers specific to bone turnover in healthy young adult men.

With ethical approval and informed consent, seventeen healthy males (age 28.2 ± 1.1 years, height 1.71 ± 0.02 m, mass 77.6 ± 2.9 kg, BMI 24.7 ± 0.8 kg m^-2; means ± S.E.M.) were screened for bone health and calcium intake. All subjects were non-smokers, not habitually active and maintained normal physical activity and dietary intake for the period of study. For five consecutive days serum samples were obtained following an overnight fast and assayed for N-MID osteocalcin (OC), a specific marker of bone formation, and C-terminal fragment of pyridinium crosslinks (CrossLaps▓trade│), specific for bone resorption, by electrochemiluminescence (Roche Diagnostics). In addition 24 h and first morning void (FMV) urinary samples were collected and analysed for pyridinium crosslinks (pyridinoline, Pyr and deoxypyridinoline, D-Pyr) and creatinine (Cr) by high performance liquid chromatography (HPLC). The individual critical difference (CDI; P < 0.05), i.e. the minimum significant difference (P < 0.05) between two measurements, was calculated for each analyte according to Fraser & Harris (1989). Changes greater than the CD are considered to be representative of a significant modulation of biological activity (Panteghini & Pagani, 1996). Pearson bivariate correlation was used to examine relationships between the dependent variables.

Analyte concentrations were found to be within the normal reference range for young healthy males. The mean critical difference for urinary analytes was found to be approximately 3-fold that of the respective serum measures of bone resorption. A low correlation between serum measures of pyridinium crosslinks and urinary measures of 24 h D-Pyr (r = 0.280, P = 0.009), 24 h Pyr (r = 0.229, P = 0.025), FMV D-Pyr (r = 0.088, P = 0.234) or FMV Pyr (r = 0.14, P = 0.122) indicates only limited agreement between related measures of bone resorption.

In conclusion, biochemical markers of bone turnover differ in their within-subject biological variance which impacts on their ability to detect change.

This work was supported by Enterprise Ireland Grant SC/228.