Current performance of C-reactive protein determination and derivation of quality specifications for its measurement uncertainty

From External Quality Assessment data, current harmonization of CRP measuring systems appears to be satisfactory, the inter-assay CV being well below 10%. The inter-method variability is even better (close to 3%) when the widely used measuring systems are compared at CRP concentrations employed as cut-off for detecting sub-clinical infection (i.e., 10.0 mg/L) and measurement variability estimated, according to ISO 20914:2019 Technical Specification, from the intermediate within-lab reproducibility of 6-month consecutive measurement data. According to the state-of-the-art model (which is better suited for CRP), the maximum allowable measurement uncertainty (MAU) for CRP measurement on clinical samples with 10.0 mg/L concentrations is 3.76% (desirable quality). As measurement uncertainty (MU) of the only available reference material (ERM-DA474/IFCC) is ∼3%, to fulfil desirable MAU on clinical samples, IVD manufacturers should work to keep the contribution of remaining MU sources (commercial calibrator and intermediate within-lab reproducibility) lower than 2.3%.