Performance characteristics of six automated 25-hydroxyvitamin D assays: Mind your 3s and 2s.

BACKGROUND Measurement of 25(OH)D has evolved rapidly, with an increased number of high-throughput automated immunoassays becoming available in recent years. The aim of this study was to fully evaluate six commercially available automated 25(OH)D immunoassays, including 2 newly available (Beckman Coulter) and one recalibrated (Siemens) assay. Comparisons were made to specifically identify the effect of the absence or presence of 25(OH)D2 on these assays. DESIGN AND METHODS Access2 and UniCel DxI 800 (Beckman Coulter), ARCHITECT i2000SR (Abbott Diagnostics), ADVIA Centaur XP (Siemens), Liaison XL (DiaSorin) and MODULAR E170 (Roche Diagnostics) assays were assessed for accuracy, imprecision, interference, limit of blank, and linearity. All were compared to an in-house LC-MS/MS method (traceable to NIST SRM 972) using Passing-Bablok regression and Bland-Altman bias plots. Method comparisons used residual serum samples with both endogenous 25(OH)D2 and 25(OH)D3 (n=50) or 25(OH)D3 only (n=86). Comparisons with all 136 samples were intended to simulate real-world laboratory testing. RESULTS The majority of assays under-recovered 25(OH)D in comparison to LC-MS/MS, with three of six immunoassays affected by the presence of 25(OH)D2. Imprecision was greatest at 25(OH)D concentrations near the decision limits used to assess deficiency. Only two of six immunoassays would meet the recommended bias criteria of <5%. CONCLUSIONS Although standardization efforts continue, these differences in performance remain a concern. Clinicians should be aware when comparing results among assays and using them to determine adequacy of 25(OH)D stores in various patient populations.