Clin Chem Lab Med. 2025 Aug 12. doi: 10.1515/cclm-2025-0554. Online ahead of print.
ABSTRACT
OBJECTIVES: Accurate 25-hydroxyvitamin D (25-(OH)D) assays are essential for defining vitamin D status and ensuring appropriate clinical decisions. Standardization efforts, including the Vitamin D Standardization Program (VDSP), aim to minimize assay variability. This study evaluates the measurement uncertainty (MU) of various 25-(OH)D assays and their ability to detect physiologically relevant changes over time.
METHODS: Seventeen pooled and eight single-donor serum samples were analyzed using two LC-MS/MS methods and 13 immunoassays, each applied in two independent laboratories. Imprecision, bias, and MU were assessed relative to the University of Ghent’s reference measurement procedure (RMP). Results were compared against analytical performance specifications (APS) from VDSP, JCTLM-TF-RMSI, and IFCC C-BM based on physiological 25-(OH)D variation. A graphical approach was introduced to visualize MU in relation to clinical relevance.
RESULTS: LC-MS/MS methods consistently met all APS criteria. Several immunoassays also achieved acceptable MU, although significant bias or inter-laboratory variability was observed for some of them. Slightly more than half of the assays met the desirable Joint Committee for Traceability in Laboratory Medicine Task Force on Reference Measurement System Implementation (JCTLM TF-RMSI) MU threshold (≤10 %), while four exceeded the minimum acceptable limit (≤15 %). The IFCC C-BM physiological approach identified a similar subset of assays. The graphical representation effectively illustrated method reliability across the tested concentration range.
CONCLUSIONS: Measurement uncertainty remains a major challenge for 25-(OH)D assays. The integration of MU-based APS and graphical visualization provides a comprehensive framework for evaluating assay performance. These findings highlight the importance of selecting assays capable of reliably detecting clinically meaningful changes in vitamin D status.
PMID:40785082 | DOI:10.1515/cclm-2025-0554