Biological variation in inflammatory and hemostatic markers

Concentrations of inflammatory and hemostatic variables are influenced by biological variation, which is the natural within-subject variation over time.

The aim of this study was to determine fibrinogen, C-reactive protein (CRP), platelet aggregation, thrombin generation and prothrombin time (PT): (i) the number of repeated measurements needed to obtain the true habitual concentration of an individual; (ii) the recommended analytical imprecision for diagnosis and monitoring; (iii) the recommended analytical bias; (iv) the contribution of analytical imprecision to test result variability; (v) the index of individuality; (vi) the reference change value; and (vii) the seasonal variation.

We collected 520 blood samples over a 1-year period from 40 healthy individuals, and determined the between-subject, within-subject and seasonal variation in fibrinogen, CRP, platelet aggregation, thrombin generation and PT.

One or two repeated measurements were sufficient to establish the true habitual concentration, except for platelet aggregation and peak thrombin generation, where at least four and nine repeated measurements were needed, respectively. For diagnosis, the maximal recommended coefficient of analytical variation (CV) was 4%-27%, except for CRP (77.7%). For monitoring, these CVs were on average 3% lower. Recommended analytical bias varied between 1.7% and 33.2%. Finally, seasonal variation was observed in concentrations of fibrinogen and thrombin generation, which could explain approximately 11% of their total variation.

This study provides insights into the biological variability of selected inflammatory and hemostatic markers, which can be used for sample size calculations and to determine the analytical quality specifications for their respective assays.