HPLC Method for 25-Hydroxyvitamin D Measurement: Comparison with Contemporary Assays

BACKGROUND: The concentration of 25-hydroxyvitamin D [25(OH)D] in serum has been designated the functional indicator of vitamin D (VitD) nutritional status. Unfortunately, variability among 25(OH)D assays limits clinician ability to monitor VitD status, supplementation, and toxicity. METHODS: We developed an HPLC method that selectively measures 25-hydroxyvitamin D2 [25(OH)D2] and D3 [25(OH)D3] and compared this assay with a liquid chromatography-tandem mass spectrometry (LC-MS/MS) method, a competitive protein-binding assay (CPBA) on the Nichols Advantage platform, and an RIA from Diasorin. RESULTS: For the new HPLC assay, between-run CVs were 2.6%-4.9% for 25(OH)D3 and 3.2%-13% for 25(OH)D2; recoveries were 95%-102%; and the assay was linear from 5 microg/L to at least 200 microg/L. Comparison data were as follows: for HPLC vs LC-MS/MS, y = 1.01x - 4.82 microg/L (Sy/x = 4.93 microg/L; r = 0.996) for 25(OH)D3, and y = 0.902x - 0.566 microg/L (Sy/x = 2.56 microg/L; r = 0.9965 for 25(OH)D2; for HPLC vs Diasorin RIA, y = 0.709x - 5.86 microg/L (Sy/x = 7.35 microg/L; r = 0.7509); and for HPLC vs Nichols Advantage CPBA, y = 1.00x - 3.60 microg/L (Sy/x = 32.7 microg/L; r = 0.6823). CONCLUSIONS: The new HPLC method is reliable, robust, and has advantages compared with the Nichols Advantage CPBA and the Diasorin RIA. The Nichols Advantage CPBA overestimated or underestimated 25(OH)D concentrations predicated on the prevailing metabolite present in patients' sera.