Background: Comparing four fully automated 25-OH-D immunoassays to a commercially available liquid chromatography-tandem mass spectrometry (LC-MS/MS) method with human serum samples from Finnish population.
Methods: 400 samples were analyzed with the Liaison Total Vitamin D, the IDS-iSYS 25-Hydroxy Vitamin D, the ARCHITECT 25-OH Vitamin D, the ADVIA Centaur Vitamin D Total, and a commercially available LC-MS/MS 25-OH D (PerkinElmer) method.
Results: The Liaison method mean value (95% confidence intervals) was 65.6 nmol/L (62.6 - 68.6); the IDS-iSYS mean was 70.3 nmol/L (67.4 - 73.1); the ARCHITECT mean was 69.0 nmol/L (65.5 - 72.5); ADVIA Centaur mean was 71.6 nmol/L (68.9 - 74.3), and the LC-MS/MS mean was 82.8 nmol/L (79.4 - 86.2). The regression coefficients (r) between the LC-MS/MS and immunoassays were 0.650 for Liaison, 0.757 for IDS-iSYS, 0.721 for ARCHITECT and 0.684 for ADVIA Centaur. With the Passing-Bablok analysis, none of the immunoassays gave results equivalent to LC-MS/MS. Two of the four automated 25-OH-vitamin D assays (IDS-iSYS, ADVIA Centaur) were overall in good clinical agreement with LC-MS/MS, even though the results obtained with all compared methods were not equivalent.
Conclusions: In conclusion, in routine clinical laboratory both immunoassays and LC-MS/MS are useful for measuring 25-OH-vitamin D provided that these methods are correctly standardized and especially sample pretreatment is carefully performed.