Introduction & Objective: High prevalence of vitamin D insufficiency in Iran calls for a reliable assessment of vitamin D status for diagnostic, research as well as public health purposes. Circulating level of 25(OH)D has been generally accepted as a reliable indicator of vitamin D status. HPLC-based analyses are considered as standard methods for 25(OH)D assays. The aim of this study was to set up a precise, reliable and rather simple HPLC-based method to determine serum levels of 25(OH)D using ultra violet detector.

Materials & Methods: Serum proteins were precipitated using ethanol and then methanol: iso-propanol (90:10, v/v). 25(OH)D was then extracted using hexane, which was then evaporated under nitrogen flow. The sediment reconstituted in methanol was further filtered. Finally, 20μL of the filtrate was injected to the column, teknochroma tracer excel 150×4, 3 μm. Chromatography was run with the mobile phase of methanol:water (85:15, v/v) containing 0.01% BHT at the column temperature 40°C. The vitamers 25(OH)D3 and 25(OH)D2 were detected at 265nm. For further evaluation of the method, 90 human serum samples were analyzed for 25(OH)D using HPLC, competitive protein binding assay (CPBA) and radioimmunoassay (RIA). The degree of agreement between those three methods was evaluated using coefficient of correlation and Bland-Altman analysis.

Results: Retention times of 25(OH)D3 and 25(OH)D2 were 9.5 and 10.7 min, respectively. Standard curves of both vitamers showed linearity up to 375 nmol/L (D3) and 187.5 nmol/L (D2). Recovery percents for 25(OH)D3 and 25(OH)D2 were found 101±5% and 100.8±5.4%, respectively. Intra- and inter-assay variations for 25(OH)D3 were 8.1% and 12.6%, respectively. The results of RIA and CPBA were significantly higher than those of HPLC (p=0.02). RIA, compared to CPBA, showed more agreement with HPLC.

Conclusion: The new HPLC method for serum 25(OH)D determination is reliable and rather fast with the advantage of detecting both 25(OH)D2 and 25(OH)D3.