A critical appraisal of four Automated Immunoassay Tests for the Measurement of 25-Hydroxyvitamin D

Abstract

Background: Vitamin D deficiency is one of the major health issues worldwide, therefore a reliable routine assay to assess vitamin D status in the circulation is required. Currently, total 25-Hydroxyvitamin D (25(OH)D), the sum of 25(OH)D2 and 25(OH)D3, is considered as the most accurate indicator for vitamin D status. However, patients on vitamin D therapy should be monitored by measuring 25(OH)D2 and 25(OH)D3 separately. The concentration of total 25(OH)D can be measured by various methods including liquid chromatography-tandem mass spectrometry (LC/MS-MS) and immunoassay. Aims: To evaluate and compare the ability of measuring 25(OH)D by LC/MS-MS and immunoassay. Also, to critically assess the performance of four common automated immunoassays Elecsys, Centaur, iSYS, and Architect assays of measuring the total 25(OH)D, compared to LC/MS-MS analysis. Method: An extended literature review, searching of various keywords from 2015 to 2020. Results: LC/MS-MS can measure 25(OH)D2 and 25(OH)D3 separately with a high level of analytical performance. Coefficient of variation (CV%) values of precision are <5.03%. The bias of accuracy ranges (0.3-2.7%) and the Lower limit of Quantification (LQ) is 6.25 nmol/L for each. However, automated immunoassays capture 25(OH)D2 and 25(OH)D3 and calculate the total without distinguishing between them, with %CV values of precision of 19.1 to 1.0%. The bias of accuracy ranges from 0.6 to 4.1%. The LQ values of automated immunoassays varies between 7.5 to 17.5 nmol/L of total 25(OH)D. Furthermore, the Architect and iSYS assays show good agreement with LC/MS-MS, while the Elecsys and Centaur assays show high biases against LC/MS-MS. Conclusion: The LC/MS-MS method shows the most accurate performance for measuring 25(OH)D, with high level of sensitivity, although it has a limitation of separating some metabolites leading to over-estimation 25(OH)D level; some approaches can overcome that. Automated immunoassays perform a different agreement with LC/MS-MS causing a variety of classifying patients, therefore, clinicians should consider the utilised assay before making clinical decisions.