Conclusions
All immunoassays can be used in routine 25(OH)D measurements, still fairly diagnosing patients' status. Recent standardization attempts seem not to contribute too much to clinical diagnosis. A clinical laboratory must at least be aware of its method to avoid misinterpretation of results.
Methods
Beckman immunoassays were assessed for imprecision, accuracy, limit of blank (LoB), limit of detection (LoD), limit of quantitation (LoQ), linearity, interference, and carryover. One hundred and nineteen samples were run on DxI 800, Access2, and E 170, and agreement with the LC-MS/MS method was evaluated.
Results
DxI 800 and Access2 assays showed good performances in terms of LoB, LoD, LoQ, linearity, and interference. All immunoassays showed negative biases ranging from - 8.6% (DxI 800) to - 19.2% (Access2). DxI 800 and Access2 systems had proportional biases, and the E170 system had a constant bias with the largest random error. Concordance correlation coefficient values ranged from 0.941 (CI: 0.917-0.958) for DxI800 to 0.854 (CI: 0.811-0.889) for Access2. Kappa (κ) coefficients were found moderate for Dxl (0.709; CI: 0.581-0.837) and E170 (0.771; CI: 0.587-0.844) and fair for Access2 (0.572; CI: 0.428-0.716). Conclusions: All immunoassays can be used in routine 25(OH)D measurements, still fairly diagnosing patients' status. Recent standardization attempts seem not to contribute too much to clinical diagnosis. A clinical laboratory must at least be aware of its method to avoid misinterpretation of results.