Evaluation of the new restandardized 25-hydroxyvitamin D assay on the iSYS platform.

ABSTRACT

BACKGROUND: 25-hydroxyvitamin D [25(OH)D] is the most reliable biomarker of vitamin D status, but until now 25(OH)D assays have suffered from inter-laboratory and inter-assay discrepancies. In the setting of the international Vitamin D Standardization Program, Immunodiagnostic Systems (IDS) recently reformulated and restandardized the 25(OH)D immunoassay available on the automated iSYS platform. In the present study, we evaluated this new generation of the 25(OH)D immunoassay (IS-2500). METHODS: Repeatability and within-laboratory imprecision were verified according to the Clinical and Laboratory Standards Institute EP15-A3. Results from the sera of 63 patients were compared with those obtained with the previous iSYS method (IS-2700S) using Passing-Bablok and Bland-Altman analysis. The prevalence and bias-adjusted kappa (PABAK) coefficient was calculated to assess the agreement of vitamin D status provided by the two iSYS immunoassays. Fourteen Vitamin D External Quality Assessment Scheme (DEQAS) samples were used to evaluate inaccuracy. RESULTS: Using the EP15-A3 protocol, repeatability and within-laboratory imprecision obtained with the new iSYS method were lower than 6% and 8%, respectively. These results are consistent with the manufacturer's claims. In more adverse conditions (50 measurements over 15days with multiple calibrations), the within-laboratory imprecision was 14.8% (39nmol/L) and 7.7% (155nmol/L). 25(OH)D concentrations measured with the new assay showed a strong correlation with those provided by the previous version (r=0.969, p<0.0001). The Passing-Bablok regression equation was as follows new assay=1.079 x (previous assay) - 3.6nmol/L. The PABAK coefficient of 0.810 reflected almost perfect agreement between the two immunoassays to classify patients according to their vitamin D status (85.7% of agreement). Using DEQAS samples, the mean inaccuracy bias was lower than 5% when the new iSYS method was compared with LC-MS/MS methods and the NIST reference measurement procedure. CONCLUSION: The new generation of the iSYS immunoassay evaluated in this study meets requirements for routinely measuring 25(OH)D levels in clinical laboratories.