Analytical Bias in the Measurement of Serum 25-Hydroxyvitamin D Concentrations Impairs Assessment of Vitamin D Status in Clinical and Research Settings

• Published in: PloS one Vol. 10; no. 8; p. e0135478
• Main Authors: Black, Lucinda J., Anderson, Denise, Clarke, Michael W., Ponsonby, Anne-Louise, Lucas, Robyn M.
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 12.08.2015; Public Library of Science (PLoS)
• Subjects: 25-Hydroxyvitamin D; Accuracy; Adolescent; Adult; Assaying; Calcifediol - blood; Chromatography; Chromatography, Liquid; Dietary supplements; Epidemiology; Ethics; Female; Health services; Humans; International standards; Laboratories; Liquid chromatography; Male; Mass spectrometry; Mass spectroscopy; Measurement; Medical laboratories; Medical research; Metabolites; Middle Aged; Multiple sclerosis; Quality assurance; Reproducibility of Results; Spectroscopy; Standardization; Tandem Mass Spectrometry; Vitamin D; Vitamin D - analogs & derivatives; Vitamin D - blood; Vitamin D Deficiency - blood; Vitamin D Deficiency - diagnosis; Vitamin deficiency; Young Adult; Australia; Western Australia Australia; Tasmania Australia
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0135478

Measured serum 25-hydroxyvitamin D concentrations vary depending on the type of assay used and the specific laboratory undertaking the analysis, impairing the accurate assessment of vitamin D status. We investigated differences in serum 25-hydroxyvitamin D concentrations measured at three laboratories (laboratories A and B using an assay based on liquid chromatography-tandem mass spectrometry and laboratory C using a DiaSorin Liaison assay), against a laboratory using an assay based on liquid chromatography-tandem mass spectrometry that is certified to the standard reference method developed by the National Institute of Standards and Technology and Ghent University (referred to as the 'certified laboratory'). Separate aliquots from the same original serum sample for a subset of 50 participants from the Ausimmune Study were analysed at the four laboratories. Bland-Altman plots were used to visually check agreement between each laboratory against the certified laboratory. Compared with the certified laboratory, serum 25-hydroxyvitamin D concentrations were on average 12.4 nmol/L higher at laboratory A (95% limits of agreement: -17.8,42.6); 12.8 nmol/L higher at laboratory B (95% limits of agreement: 0.8,24.8); and 10.6 nmol/L lower at laboratory C (95% limits of agreement: -48.4,27.1). The prevalence of vitamin D deficiency (defined here as 25-hydroxyvitamin D <50 nmol/L) was 24%, 16%, 12% and 41% at the certified laboratory, and laboratories A, B, and C, respectively. Our results demonstrate considerable differences in the measurement of 25-hydroxyvitamin D concentrations compared with a certified laboratory, even between laboratories using assays based on liquid chromatography-tandem mass spectrometry, which is often considered the gold-standard assay. To ensure accurate and reliable measurement of serum 25-hydroxyvitamin D concentrations, all laboratories should use an accuracy-based quality assurance system and, ideally, comply with international standardisation efforts.